Notes Week Ending 6/28

6/22: Of the 184 active players with at least 500 games played at the end of 2024, Mike Trout didn’t rank in the top 20 in RBI per plate appearance, despite his impeccable record when we only look at Trout. I’ll let you analyze the likely reasons. He ranked 22nd. Among the players ahead of him: Josh Naylor, Gary Sanchez, Adolis Garcia, Brandon Lowe, and Salvador Perez.

Outside of his rookie year, Trout has only led off 31 times, so that’s not “it” or even “part of the it.”

I had to get my bearings on this, so to make sure I’m not making a grievous mistake, consider the up-to-date career statistics of

Salvador Perez: 955 RBI in 6567 PA

Trout: 982 RBI in 6858 PA

6/23 N1: Without taking a really “micro” view, I can look at the NL and AL standings again and again and not see how the NL only has a three-game composite edge. The NL leads NL-AL matchups only 171-168, something which has to be reflected in the overall record for each league, where the NL is 3 games over .500 (584-581), and the AL 3 games under .500.

Even with the 25-53 White Sox, the AL seems to be hanging in there cumulatively because it doesn’t have the 18-60 Rockies. The AL also has one fewer team 10+ games under .500, and I suppose that subtly goes a long way. The bad teams in the AL are the Orioles, Athletics, and White Sox; in the NL, the Nationals, Marlins, Pirates, and Rockies.

If you just focus on the good teams and the playoff races, as everyone understandably is, the AL is uninspiring. The three NL teams currently in the wild card places, and the first team on the outside looking in, combine for an average winning percentage of .563; the equivalent group in the AL is at .527.

It could also be the case that the AL were top-heavy, and that would be why there isn’t depth of teams on pace for 90+ wins. But the NL has the edge with its division winners as well, as they are playing at a .603 pace. The Tigers, Yankees, and Astros have an average winning percentage of .594.

It is all very strange, but the takeaway seems to be that the AL standings are ugly but not the picture of a bad league.

I have to say that subjectively, I still believe the NL is quite a bit better. Of the two sleeper/underachieving teams, even with the Sale injury, I prefer the Braves to the Orioles. I also think there is something to say for the Nationals, Marlins, and Pirates, while I think the Angels will probably collapse. A conservative prediction (i.e., one likely to be right) is that the AL will make up some ground with its wild card teams by season’s end, but lose the interleague battle by more than the 7 or so games for which it is currently on a pace.

6/23 N2: There are 14 current E.R.A. qualifiers who have allowed at least a hit an inning. Except for Brandon Pfaadt and Bailey Ober, who are giving up more than 10 hits-per-9 innings, there isn’t a great deal of variation among their hit rates, as a pitcher can generally only be so far over 9 and still be effective and efficient enough to be an E.R.A. qualifier. Along the same lines only two of the pitchers have E.R.A.s over 5.00, but only Chris Bassitt (a 3.61 E.R.A.) and Tomoyuki Sugano (3.55) in the group have E.R.A.s under 4.00. (I initially didn’t see Sugano, who has given up 84 hits in 83.2 IP, but I’m not going to let him cancel my post.)

Bassitt has the 9th-worst hit ratio among E.R.A. qualifiers. After Sugano, you have to go to Jose Soriano, whose 8.6 hits-per-9 is 20th-worst, to find the next pitcher with an E.R.A. better than Bassitt’s (Soriano’s is 3.39). Soriano has managed that despite having walked 43 to Bassitt’s 20 even though their innings counts are within 1 of each other. Soriano’s secret has been his low home run rate. He’s given up 1 home run for every 31 innings. Jose Ramirez hit 2 home runs against him on April 4, and Andrew McCutchen hit 1 against him on April 22, but that’s been it.

Having allowed 10 home runs, Bassitt can’t claim anything like that. What stands out instead is that he’s allowing just a .207 RISP average, compared to .260 overall. With runners on first base only, hitters are hitting .305 against him, but he hasn’t given up any home runs, and 16 of the 18 hits have been singles. So there’s a decent chance Bassitt hasn’t given up any immediate runs (or RBI) in the runner-on-1st-only situation. And then we know that once the runners have graduated to RISP status, he’s been stingy.

6/23 N3: I didn’t know if it had anything to do with why he’s overachieved relative to his hit rate, but another thing I noted about Bassitt’s situational stats is that not only is his BA allowed with runners in scoring position excellent, but so is his OBP allowed of .267. He’s actually hit 3 batters with RISP, but what goes into that is just 4 walks allowed in the 90 matchups. My belief was that the walk rate with RISP, even throwing intentional walks out, is higher league wide than without RISP, and that these data therefore made Bassitt unusual.

To get the numbers, and with a group that would reflect starters and not relievers, I totaled up the E.R.A. qualifiers in non-RISP and RISP situations. I found that their combined walk rate was 6.98% in non-RISP plate appearances, and 8.27% in RISP plate appearances (intentional walks are taken out here). Bassitt’s split, on the other hand, is 5.61% non-RISP, 4.44% RISP. So his differential is extremely close to opposite.

Looking player by player among the qualifiers, however, I learned that Bassitt is less than exceptional. This could just simply be the case, or it could be that the variety in the numbers indicates the effect of the small sample size. But explaining why Bassitt doesn’t stand out, first, there are pitchers walking less than Bassitt with RISP. And second, there are just pitchers whose comparison slants more in the direction of fewer walks with RISP.

Among E.R.A. qualifiers, four pitchers have just issued one RISP non-intentional walk this season: Logan Webb, Tomoyuki Sugano, Zack Littell, and Bryan Woo. Webb’s performance really stands out, as he’s faced 92 batters with RISP, while the other three pitchers have batters faced in the 60s with RISP. While Webb has actually walked a higher percentage without men in scoring position (6.0%) than Bassitt, all four rate as good control pitchers generally. Littell has the best walk rate in the game this year when men aren’t in scoring position, and Woo also has a walk rate under 5.0% in non-RISP situations.

Webb has the third-largest “reverse split” by percentage difference, although he rates number one if we figure by ratio. The number one, two, and four cases have not been control pitchers this year, but guys who have “upped their game” with runners in scoring position. Note Luis Castillo (3.4% RISP, 9.2% non-RISP), Framber Valdez (4.4% RISP, 9.8% non-RISP), and Chad Patrick (3.9% RISP, 8.5% non-RISP). With divergences a bit smaller but still in the same family, we could put Carlos Rodon (5.8% RISP, 9.3% non-RISP) and Dean Kremer (3.5% RISP, 7.0% non-RISP).

The strongest trend in the other direction belongs to Freddy Peralta. Peralta leads E.R.A. qualifiers in RISP walks (14) and walk percentage (20.6). As his control is actually slightly above-average without RISP (6.8%), you can bet he leads in differential between RISP percentage and non-RISP percentage, too (if we go by ratio, nothing changes).

There are 10 other pitches with a RISP walk percentage 5% higher than their non-RISP percentage, and I would classify them like this.

Going just by the data, Yusei Kikuchi, Randy Vasquez, and Erick Fedde are bad control pitchers whose walk tendency becomes all the worse with RISP.

Then Jacob deGrom, Joe Ryan, and Zack Wheeler have top-15 walk rates when guys aren’t in scoring position, but they walk guys when there are guys in scoring position. Their average walk rate with RISP is 10.6%.

That needs a footnote with Wheeler, however. Wheeler has been amazing this year in just not having guys in scoring position, very much unlike Webb. Only 5 RISP walks go into Wheeler’s technically elevated percentage. His percentage of his total walks with RISP, 22.7%, is actually lower than the 23.0% for the dataset as a whole. Wheeler has had 41 plate appearances compiled against him with runners in scoring position; behind him among E.R.A. qualifiers, you have to go all the way up to deGrom and his 52 RISP BF, along with the same 52 for Jameson Taillon. DeGrom also has only 5 walks with runners in scoring position. So we don’t want to make much of the possibility that these guys are nibbling in that situation.

To continue with the next group of large walk differential pitchers in the normal direction — Jesus Luzardo, Tyler Anderson, and Andre Pallante don’t rate with the last three pitchers mentioned in their non-RISP walk rate, but their high rates with RISP (between 12.4% and 13.6%) were isolated to that situation nonetheless. In fact, they rated 20th to 25th in non-RISP PA of 73 qualifiers by walk rate. None have the Wheeler excuse, with RISP walks in double digits.

For the final group, there are two pitchers I couldn’t classify, two pitchers who you could say have middling walks without RISP but high walk rates with RISP. These are Jeffrey Springs and Andrew Heany.

So, back to Bassitt. His 1.17% lower wak rate in RISP situations rated him just 26th of 73. He was joined by five other pitchers with a better walk rate in RISP than in non-RISP situations, making for 31 in all. The respective numbers of 42 pitchers with a better walk rate in non-RISP, versus 31 with a better walk rate in RISP, seem to me to go right with the average 1.29% differential.

As I suggested, I think my first reaction to Bassitt’s walk rate with RISP both didn’t sufficiently recognize the stingy baseline his general walk rate set and how variable the rates typically are, and how many pitchers could therefore be found that have a larger reverse split. However, based on some other observation, I do feel fairly confident there are pitchers who consistently have larger differentials between their RISP walk percentage and non-RISP walk percentage than league average. I think there’s a strong theoretical argument for it. Pitchers construe the situation of a runner in scoring position differently, have different philosophies about how much of a premium to put on that runner in scoring position. They are stubborn to different degrees, patient to different degrees. A pitcher with one dominant out pitch that’s an offspeed pitch might also go to it extensively with runners in scoring position, and then therefore allow more walks with runners in scoring position.

Taking the other case, of the pattern that Logan Webb and Luis Castillo have shown this year, such differences could all be random. It’s not impossible a pitcher could consciously operate more aggressively with runners in scoring position than otherwise, but it’s not particularly logical. I suppose the RISP walk rate for such pitchers could contrast with other situations where for some reason they were likely to nibble, like with 2 out and nobody on., for instance.

6/23 N4: Part of calculating those walk rates for pitchers was taking out the intentional walks, so I was forced to pay attention to that column. For 55 of the 73 pitchers, I was wasting my time in so doing, as they issued no intentional walks. Then 12 of the pitchers had just one intentional walk. But four pitchers had three intentional walks. They three Angels (Kyle Hendricks, Tyler Anderson, and Yusei Kikuchi), and Dustin May.

The strange Dustin May thing aside (he is the only Dodger starter to issue an intentional walk this season), we give the identity of the pitchers only a fleeting thought, as intentional walks are decreed by the manager. So the implication was that Ron Washington is sometimes having his starters intentionally walk batters, while no other manager is really doing that.

Getting the data to make sure, and not just for E.R.A. qualifiers, 11 of the 39 starter intentional walks this season have been issued by Angels. Arizona is next with 4 starter intentional walks.

The issue obviously is not about starter versus reliever intentional walks. Managers would presumably be just as happy to have starters intentionally walks batters as relievers, if those intentional walks came late in the game. But starters are usually gone late in the game, so starter/relievr intentional walks represents the innings dichotomy.

The exact count of intentional walks by inning this year, along with the percentage each represents.

1: 2 (0.8%)

2: 5 (1.9%)

3: 7 (2.6%)

4: 6 (2.3%)

5: 21 (7.9%)

6: 19 (7.1%)

7: 37 (13.9%)

8: 47 (17.7%)

9: 37 (13.9%)

XI: 85 (32.0%)

It would be very strange if all of these Angel starter intentional walks came when intentional walks are normally issued, late in the game. But for the sake of due diligence, we have to check.

A first indication of where the numbers will fall is that, while the Angels lead MLB easily with 23 IBB, that leaves only 12 additional intentional walks for their relievers. Any sense of the data will tell you that starters generally are issuing way less than 48% of intentional walks, as the Angels are (in fact, the MLB starter percentage is 15%).

In the by-inning data, we basically see a spike at inning 5, inning 7, and in extra innings.

Overall, the Angels have issued 8.6% of baseball’s intentional walks. That’s 2.6 times as many as an average team, since at average team is 3.3% of the league.

Separating out each of the threshold points, the Angels’ share of league walks progressively declines. They have issued 20% of the league’s walks in innings 1-4, 17.5% of the league’s walk in innings 5-6, 5.8% of the league’s walks in innings 8-9, and 3.5% of the league’s walks in extra innings.

There is only slight indication, if any, that the Angels walk more men intentionally per inning as the game progresses. Their top inning for intentional walks is the 5th (5 IBB). In no other inning have they walked more than 3.

It stretches credulity, but could all of this just be the Aaron Judge factor? Could the Angels have walked him 10 times intentionally or something, and done that regardless of inning?

I went and found the identity of the hitters they’ve been intentionally walking early in the game. While the Angels have been responsible for four of the seven intentional walks in the first two innings, and two of those were indeed given to Judge, others were given to Jose Ramirez and Corey Seager. Then none of the Angels’ 7 intentional walks over the 4th and 5th were of Judge. Three were to Jose Ramirez, two to Vladimir Guerrero Jr., one to Cal Raleigh, and one to Tyler O’Neill.

While my goal has just been to document the difference in the Angels’ approach, for the record, yes, I do rather assume they have been proceeding foolishly. My main criticism has long been of the intentional walk, not early-in-the-game intentional walks. It’s possible that intentional walks are generally as harrmful late in the game as earlier, although extra innings are certainly a completely different thing, where potential runs in the bottom half of the inning can be meaningless. You also don’t need to worry as much about potentially falling far behind if giving up that one extra hit will also put you 2 or 3 behind with not much time to go.

Dustin May’s intentional walks have all come since June. They’ve come in separate starts. I suspect that Dave Roberts has fallen into a bit of a pattern where he’s adjusted what he’s doing without realizing it. He did the one intentional walk with May, and then he did another the next time. He’s become consistent with himself, not with the percentages. It’s hard to believe these have really been the three most compelling times all year to intentionally walk a batter with his starting pitchers, and the occasions just happened to all come in June with Dustin May on the mound. If Roberts has changed his mind about the right strategy, then he had better start intentionally walking batters with other starters, unless he thinks Mays has unique huge platoon splits (that’s an argument, come to think of it). I can understand the IBBs he issued to Juan Soto and James Wood with May. But to Mike Yastrzemski?

6/23 N5: I had wanted to mention something about the Mariners in the last note but forgot. While they are third in total intentional walks this year (starter and non-starter) with 14, behind the two LA teams, 11 of those have come in extra innings. So if you are an opponent of the intentional walk as I am, you can’t lay the statistic at the feet of Dan Wilson. The overall IBB walks are misleading, and don’t give an idea of his general fondness for the strategy. Obviously, with the stats I cited earlier, the IBB rate in extra innings is so much higher than otherwise, how many extra innings a team plays has a lot to do with how many intentional walks it has.

6/23 N6: David Nemec, author of The Beer & Whiskey League, excels in creative vocabulary. Example: he calls a season’s opener the “lidlifter.”

6/24 N1: Excited for Chase Burns’ debut tonight. Has gone 7 innings a couple of times recently, but the most pitches he’s thrown this season is 88.

6/24 N2: The stolen base success rate this month is low enough we have to take notice — it’s 73.9%. Last year, for the whole season, it was 79.0%, and in 2023, the first year of the new rules, it was 80.2%. From 2020-2022, it was just a bit above 75% each year, so this month’s success rate is well off that. It is just one month, but it certainly seems backward that runners are succeeding less often than before the rul change. Success in March/April (78.4%) and May (76.7%) was much higher than it’s been this month, but we did get indication the going would be tougher than it was in 2023-2024.

I can’t compare the size of the two changes offhand, but attempts are also down this month. I looked at successful steals a month for all months from 2023 through this month, instead of calculating attempts as I should have, but I can say that this year attempts by month have gone from 1.033 in March/April to 0.950 in May to 0.848 in June. So that divorces actual stolen bases from the lower success rate, and shows a separate decline.

Just as, if you look at intervals of horse races and I think of human races, an even clip is not expected, the example of 2024 does indicate that monthly stolen bases may typically have a u shape. As stolen bases have to do with effort in a way that hitting doesn’t, it makes sense that everyone is eager early on, but then the reality of the long season sets in, and teams pull back. Then maybe steals increase again later in the year when, first, the defenses themselves get tired, and then when some kids come up from the minors, and then when teams out of the pennant race are more willing to experiment, on the theory that “they don’t have anything to lose.”

But the data of 2025 is still shaping up as very different from 2024.

Successful steals per game in the first three months this year

March/April 0.810

May 0.729

June 0.626

Successful steals per game by month 2024

March/April 0.743

May 0.726

June 0.693

July 0.737

August 0.791

Sept/Oct 0.775

Last year’s “June swoon” was modest at most, with the August surge standing out more. So, while not much should generally be made of statistical fluctuation, I am not confident we will see a bounce back, because the magnitude of this decline does not have recent precedent.

During 2023, my reading is that teams were trying to figure out what was going on, and how much of a role stealing should play in their arsenal. I do not then think the monthly data for the year are particu;arly instructive. The monthly rates of successful steals were flat through August, with a low of 0.687 (July) and a high of 0.740 (June). September/October surpassed any of the earlier months, at 0.781.

The month isn’t over, and that’s important to note. That we would see an aberration in this shortened month is therefore more likely than if it were a complete month. But as it stands, if you compare the 0.626 steals this month to the 0.730 average of the 15 months over the three years with the new rules, it’s the highest disparity from the average, negative or positive.

Also interesting is that March/April of this year was the month with the most steals. So it’s been a striking difference, a decrease in attempts of 18%, and a decrease in successful steals of 23%. One would think months close together in time would tend to trend together. The pattern is of abandonment.

Because the overall decline is large enough in size and apparent beyond just this year’s teams, I trust it is not wholly illusory, but I do know that stolen bases are a positively skewed statistic where a few teams and even players can have an impact. Because of that, I wanted to analyze the change this year on a team-by-team basis.

I found that 21 of the 30 teams have attempted fewer steals per game this month tthan they did over the first two months. I suppose that’s not an overwhelming difference — not a trend like the one of skyrocketing strikeouts 10 years ago ago or something.

The Reds attempted 74 steals in the first 59 games, but have a rate of going just a little over half of that in June (13 attempts in 20 games). That makes for the first biggest fall-off by number of steal attempts-per-game.

We can put the Blue Jays and Dodgers together. Combining for 93 attempts in 116 games through May, they didn’t loom large on the stolen base landscape, but this month, they are baseball’s two most stationary units. The Blue Jays have just 6 steal attempts in 19 games; the Dodgers, 8 attempys in 21 games.

It’s easy to see Ohtani in that. Perhaps because he’s turned his mind to pitching or has been worried about injury, he hasn’t atttempted to steal a base this month. At the end of April, he seemed to have just resumed where he left off in 2024; his 9 steals placed him top 10 in MLB.

Atlanta, which has gone from 0.789 steal attempts a game to 0.400, belongs with Toronto and the Dodgers.

The Rays have the most attempts per game on the year (1.628) with the Brewers next (1.544). Both teams have remained active this month (1.450 for the Rays and 1.25 for the Brewers), but both have fallen off. However, with Tampa Bay, even though their good June hitting has maybe supplied them with lots of opportunities, the loss of Chandler Simpson could just be the explanation for the fall-off. They still have run more in June than any other team.

The Brewers have been passed this month in stolen base attempts per game by the Mariners (1.350). Sixth in attempts per game in March/April and March (1.246), the Mariners do show absolute increase.

As is typical, Houston (from 0.707 to 1.150) and the Angels (from 0.579 to 0.905) have zagged where everyone else has zigged.

The comparison is a bit apples to oranges, and yes, we can play games with numbers. But the success rate decline is more striking to me than the fall-off in attempts, from the perspective given by the following stat.

Over all of 2024, this month’s 73.9% success rate was bettered by 26 of the 30 teams. The only exceptions were the White Sox, Rockies, and Angels, with the Twins at 73.9% exactly (interesting the correlation between bad teams and low success rate this list happens to suggest). The possibility that the game is changing, that the pitchers and defense have caught up to the runners, is brought to the fore by that framing. A 74% success rate wasn’t something we saw last season.

I will preface this by saying that my conviction on the matter is stronger than the evidence warrants. I guess I am bringing a lot of psychology into the question, but I do flatter myself that sometimes I have good intuition about psychology. I think what has happened with the recent trend of fewer attempts is that the runners thrown out early in the season threw cold water on stealing. I think the decrease in attempts is therefore simple conditioning. I am blanking on a good example with stolen bases, but there aren’t a lot of Jose Altuves, who make outs trying to take extra bases, and just keep doing it. It is natural to become discouraged, or at least more cautious.

The theory that teams have already number crunched the success rate and issued a directive to change course is also an interesting possibility. That would be making a change pretty much in real time.

6/24 N3: Boy, that Cal Raleigh is consistent. Home runs totals since 2022 of 27, 30, 34, and 32….Wait, we’re not even at the halfway mark! If we take stats at face value as evidence of improvement, remarkable that one of the game’s most improved players was starting at such a high level.

6/24 N4: I consider the Diamondbacks very much “out of sight, out of mind” and to be a disappointment after providing early season intrigue. So I don’t know if I’m more surprised to find that they are 13-7 this month, after going to the record, or second in runs scored this month.

They have a .462 slugging average, and are second in June slugging average, too. To Colorado (.468)! Somehow, with a June OPS that is also 4th in MLB, Colorado is just in a tie for 17th in runs scored.

6/24 N5: If a group of your friends can’t get into a contest to pick the shortstop who will finish with the best slugging avereage this year, I say there is no hope for you, and no hope for baseball. It’s a wide open contest. Strong cases can be made for EDLC (currently .513), Zach Neto (.498), Bobby Witt (.490), CJ Abrams (.487), Francisco Lindor (.474), and even Trea Turner (.450) and Gunnar Henderson (.433). I have no idea what to do with Jacob Wilson (.493). I assume Jeremy Pena (also .493) will fall back, although if he turns more toward power and thinks less about average, maybe not. At .391, unfortunately, it’s probably too late for Mookie Betts. I’ll also put Bo Bichette, Dansby Swanson, Anthony Volpe, and Carlos Correa as part of the mutuel field, but maybe one of them will come back and bite me.

6/24 N6: I’m not a fan of “national days” like “national tight end day,” but if declaring a “national oblique day” would solve the problem, I’d be more than up for it.

6/24 N7: I just was rather shocked to look in a Stathead Statistical Filter and to see RBI in the list as “Runs Batted In” (what’s that?). I tend to be very proper, but I’m not sure I’ve called it that in recent memory. For some reason, RBI seems to engender the laziness of a government acronym. I understand the crutch with some of the ones of government or policy or law, because they can reach to five or six letters.

6/24 N8: In the process of some research which will seem less random when I eventually post on it, I learned that there was an RBI on 291 of the 697 triples last year, or 41.75%. I was originally surprised that the percentage was so low, that so many triples guarantee nothing. But then I thought of my knowledge that only about 40% of home runs are hit with somebody on base. If the same were true with triples, then 60% of triples could not have an RBI, Conversely, if somebody is on base, unless there is a scoring error of monumental proportions, it seemed to me, a triple is by definition an RBI triple. Checking, I indeed found that the number of triples with “men on” last year to be 291, exactly matching the number of RBI triples.

I then began examining the comparison of the triple data to the home run data in different light. If the percentage of home runs with no one on is 60%, as I had simplified it to myself, then, I thought, maybe that 41.75% of triples with men on actually represents a higher percentage of triples than home runs with men on?

If that proved true, then what could be the explanation?

Well, I thought of one. There is the case where a runner is on first and the batter ostensibly triples. It seems to me that some of these are gift triples. They are at the discretion of the official scorer, because the defense’s attention is on the man running home. We don’t really know whether the player would or would not have made it to 3rd or even attempted it, but with only one guy to please, the hitter, a sort of benevolent home scoring takes over whereever the game is played, and the hit is scored a triple. Of course, one could argue that some triples are taken away because the official scorer errs on the side of caution and doesn’t want to assume the runner would have made third. (When you assume you make an ass of u and me, and all that.) But the point is that this play is very much not just the recording of an event. It is not like a home run, which as it long as it goes over the fence, has no ambiguity.

When I checked the percentage of home runs with men on last year, though, it turned out it was not exactly 40%, and the small deviation here makes all the difference about the direction of the triple bias. It was 43.37%. This suggested that perhaps triples are less likely with a man on first.

We’re just getting started. Maybe 2024 wasn’t a typical year. A multiple year sample size could be important because triples are such a rare event, and we are, after all, breaking them down still further into bases-empty triples and men-on triples.

So I looked at 2005-2025 inclusive. I figure that ought to be a big enough sample size. And the data reversed. While in 2024, the percentage of “men on home runs” led the percentage of “men on triples,” 43.37 to 41.75, from 2005-2024, the verdict was in favor of triples, 43.68% to 41.98%.

My argument, however, only had to do with the “man on 1st” situation. I was just relying on that situation to increase the relative percentage of triples in all “men on” situations. So, for a purer reading, I looked at the “1st only” split Baseball Reference provides.

And here the analysis took another turn. Still using 2005-2025, I found that 19.13% of home runs were hit with a man on first only, compared to 16.81% of triples.

That’s the biggest difference we’ve found yet, and not in the direction I expected.

It nagged me, however, that maybe home run hitters just hit with a man on first a disproportionate percentage of the time, while potential triple hits are randomly distributed. I couldn’t really explain why home run hitters would come up more relatively with a “man on first only” than with men in scoring position, and I knew that somewhere in the on-base categories, triple hitters were showing more elevation than home run hitters, but this thought did drive home to me the importance of identifying how well home run hitters were really performing and how well triple hitters, rather than just comparing the percentages.

To be specific, the percentage of hits to at-bats can be computed and can be compared in each of the situations.

So from this I saw:

Home runs (using AB, not plate appearances ) have a percentage of 3.22 with no one on, 3.32 with a man on 1st only, and 3.01 in other “men on” situations.

By comparison, triples have a percentage of 0.49 with no one, 0.46 with a man on 1st only, and 0.56 in other “men on” situations.

Takeaways?

First, there does not appear to be any bias towards more doubles/triples getting coded as triples with a man on 1st. If anything, the bias appears to be in the other direction.

Second, I don’t know what to make of the 525 more triples that were hit with men on but not a runner on first only than would be expected from the “nobody on” triple rate. Without running the numbers, I don’t know if that difference, as impressive as it seems when the enormous number of at-bats is invoked, is robust from sampling error or not.

But regardless of the statistical significance that emerges, I don’t understand why runners on 2nd and 3rd would produce official scoring or defense that would augment triples. Those runners seem immaterial.

Other theories for the elevated percentage in this situation….

(1) We could look at how triples correspond with number of outs, and how number of outs correspond with base situations.

(2) The presumably greater number of triples with men in scoring position could just reflect that hitting in general is easier there. We don’t normally think of triples as a real hitting category, but it’s “doubles plus,” really, so it is one, just with lower counts. And we must always keep in mind the sacrifice fly bias, which increase percentages of all offensive events with a runner on third and less than 2 outs, when you use at-bats as the denominator, as I did.

The home run data aren’t surprising overall, and I am particularly happy in this case to see data that make sense. Home runs go up with a “man on 1st only” probably for several reasons, but a big one is just that teams hopefully gear their lineup so they have punch-and-judy hitters hitting with no one on, which in turn means more home run hitters hitting with men on. In other words, that comparison of the 3.32% home runs with a man on 1st only, versus 3.22% with no one on, is not a straight comparison. This is a cross-sectional analysis, and if we actually looked and saw who was hitting in each situation, we would find it was not the same guys. The “no one on” situation is weighted with hitters of less power.

But the 3.01% tells us that bias of better hitters hitting with men on base is outweighed by other things when men get in scoring position. It seems reasonable that pitchers tighten the screws and are extra careful and cognizant of the home run, and the lower percentage tells us this is indeed something they can do to effect. When men get in scoring position and home run hitters are up, managers might also summon relievers who are more effective than starters. So that, yes, more powerful hitters are hitting, but they are facing better pitchers, too.

6/25 N1: I guess another reason why home run percentage could decrease with runners in scoring position would be if “base hit” mentality took hold. In other words, home run percentage might decrease because batters, to some extent, just try to get that runner in from 2nd or 3rd and don’t worry about hitting the ball over the fence. This would have to be evaluated in terms of whether these other hits did in fact go up with RISP, or at least went down less than home runs (any increase would need to hold with sacrifice flies counted as outs and added to at-bats). It’s a hypothesis that certainly fits the triple data, which are, after all, another kind of hit.

But I don’t really see evidence of RISP “base hit” mentality in the broader conversation, in the broader conversation is to some extent relevant for how the game is played. For instance, we keep talking about OPS with runners in scoring position, not batting average. In last Saturday’s Yankee game, Joe Girardi was going on and on about how the Yankees problem is they’ve recently had a terrible problem with their “slug” with runners in scoring position. A sentence that would deserve an exclamation mark, except that I find those work after long sentences. You can love him to death, but Joe Girardi still grates.

6/25 N2: In the SABR Dale Murphy interview, it came up that Murphy only hit .203 against Vida Blue. As there was no interleague play, he only faced Blue when Blue was with the Giants. While his outings were maybe a tad shorter and any claim he had to domination perhaps a bit less plausible, from some perspectives these were actually very good years for Blue. He came to the Giants in ‘78 and made the All-Star team, and followed with All-Star honors in ‘80 and ‘81 as well. In those years, he had E.R.A.s of 2.79, 2.97, and 2.45. Coming back to the team, he even retired on an up note for the ‘86 Giants, pitching to a 108 ERA+ in 28 starts.

But curiously, when you look at his Giants time altogether on Baseball Reference, his ERA+ was just 99. While we could cite his horrific 5.01 E.R.A. of ‘79 and his 78 ERA+ in ‘85, his average ERA+ in the six seasons was still 106. So the 99 is a bit hard to see, although it comes from weighting, of course. A factor was that his best ERA+, 1981, came in the strike-shortened season.

Another thing to come out of this Murphy exchange about Vida that doesn’t quite make sense to me is that Murphy’s theory for why Blue gave him trouble was that Blue had an outstanding high fastball. I don’t know if that’s true or not, that Blue had an outstanding high fastball. but Blue’s 301 strikeouts in ‘71 were his only season of 200+ (although I guess Jim Palmer also had a high fastball and didn’t strike out hitters). And Blue generally avoided home runs, only giving up 0.71-per-9 innings over his career, which doesn’t go with pitching up. The absolute numbers certainly are easily misinterpreted, as both strikeouts and home runs have greatly increased, but Blue to me still doesn’t profile as having been a high fastball pitcher.

Murphy’s answer about Blue’s high fastball does ring true if you look at his own numbers versus Blue. Somehow, despite hitting .203, Murphy was able to hit home runs against him (if perhaps making too much of the 68 matchups). Five of Murphy’s 12 hits against Blue were home runs, and he had 1 home run for every 12 at-bats against Blue, compared to the 1 out of exactly 20 he had for his career. In light of his overall 19.3 strikeout percentage, I don’t find his 26.5 rate against Blue necssarily higher than we would expect, but it is a high percentage.

If Blue developed a reputation as a Murphy killer, it might have come from 1978, Murphy’s first full season as a regular. He struck out Murphy 7 times over 17 matchups. And Murphy’s good home run percentage against Blue was largely made during Blue’s return to San Francisco as an older pitcher, when Murphy got him 3 times in 21 AB. As tiny as the samples are, we happen to see Murphy gaining the upper hand as he peaked and Blue faded.

I don’t want to let go what I stumbled on before, that Murphy’s AB 7960 were exactly 20 his home runs (398). A 1 in 398 chance, right? The more home runs you have, the less likely is an exact multiple. F@#$%! Baseball Reference and its career leaderboards for having plate appearances alongside, and not at-bats, and not allowing me to see if Murphy has the most home runs ever for someone with an exact ratio. I had thought I was on to something.

6/25 N3: An interesting question is whether underlying metrics of plate selectivity are better predictors of future walk rate, or walk rate itself…..Let’s take the case of the Yankees’ Jasson Dominguez. His 12.1% walk rate places him 30th among 161 qualifiers, but he is in the bottom two-thirds in O-Swing (49th) and Swing (48th). It simply appears that he has gotten his walks because he sees the 7th lowest percentage of pitches in the strike zone.

Dominguez doesn’t make good contact, with the 34th-worst whiff rate of the qualifiers. Yet, somehow, not only is his walk rate very good, but his SO/BB ratio (2.3-1) is also above average.

Even though he’s seen that high percentage of pitches outside the strike zone, I still wouldn’t think if you swing as much as he does, you’d be able to get so many walks. He swings an above-average percentage at Statcast balls, Statcast strikes, and all pitches together. What good is seeing balls if you don’t take them? Yet he has those 30 walks in 246 plate appearances.

Something I have explored in the past is the idea of players who do a good job getting the walk once they get to three balls. The FanGraphs/Statcast data is aggregate, and doesn’t recognize that you need to carry it out four times in a plate appearance to get a walk. Dominguez has converted his three-ball counts into walks 13.4% more than the average player. But he’s also gotten to three balls 25.3% more often than the average player. So, again, how he’s been able to do that swinging as much as he does, I don’t know.

What is Dominguez’s minor league walk history, you might ask, in consideration of whether his rate this year is a fluke? On the one hand, we might expect more of a fall-off than he has shown from the 13% walk rate he had in the minor leagues. But on the other, in his last full season of minor league ball, before his elbow injury, he had 83 walks, and a 15.3 rate.

Some other skepticism about Dominguez is that he has really failed to build upon those 3 home runs he had against the Athletics May 9. He has just one home run since, and just 6 on the year.

6/25 N4: Then there is Kerry Carpenter, whose FanGraphs pitch-level data doesn’t seem so bad that he should have the very worst walk rate among qualifiers (7 BBs in 270 plate appearances). Consider

41% of qualifiers have seen a lower percentage of Statcast strikes than Carpenter.

He’s swung the 14th most frequently of the 161 players, and has swung the 16th most frequently at pitches outside the zone.

I would say these data at most just give a hint that he could be the worst by walk rate, although that Dominguez key of “strikes seen” seems not so important in this case if we are to maintain that.

If we take the approach of focusing on his three-ball counts for additional insight, he’s reached 3 balls 64% as often as the average player. Then he’s gotten from three balls to four balls 47.5% as often as he should have.

The statistics of apportioning blame in this framework are actually fairly complicated, and all I am really comfortable asserting is that he has failed to do either half well. While the 64% is a larger number than the 48%, because it is also the larger category, it indicates that Carpenter has already fallen 19 behind the average player in terms of plate appearances where he still has the possibility of a walk. If he could have converted at the MLB average of 43.4%, however, he could have cut his walk deficiency from -19 to -8.3. But because he converted at just 20.6%, his walks and not just his 3-ball counts have been in arrears,16 less than expected.

Taking out intentional walks, Carpenter had a walk rate of 6.4% in 2023 and a walk rate of 7.1% in 2024. Less extreme percentages, obviously, making us even stronger in the expectation regression to the mean will occur. But what is discouraging is that, at least in 2024, his FanGraphs data were also all better (O-Swing of 31.3% vs. 34.6 this year; Z-Swing of 70.1% vs. 73.6% this year; while pitchers were in the zone 49.1%, vs. 52.3% this year). I have little familiarity, and can only eye ball the data, but it seems for both years Carpenter is a guy who walks less often that the FanGraphs data indicate he should. When those numbers have gotten quite bad, as they have this year, then his walk rate is particularly miserable.

6/25 N5: Fourteen straight games in which Ronald Acuna has reached base at least twice. Hard to gauge offhand, but seems marvelous.

6/25 N6: While he was just a .260 career minor league hitter, Addison Barger certainly flashed at times. He reached AAA at the age of 22 in 2022, and hit .308 with 33 2B and 26 HR across three levels that year. While he was awful at the plate in 69 games with the Blue Jays last year, his OBP in AAA in 57 games was .391. Adding to the idea that this is a guy with real athletic ability, he leads all infielders in both “max” and “overall” throwing velocity this year. But like a woman with good looks, what most really care about is that .489 slugging average he has in 2025, a product of 16 doubles and 8 home runs in just 178 at-bats, and all arguments for him are really subsidiary to that.

6/25 N7: Beyond his 0.914 WHIP, better than the 0.955 mark that led the National League last year, here are a couple of possible reasons Zack Wheeler’s number of men in scoring position have been so extraordinarily low.

With nobody out, the OBP against him is .167 (a .145 batting average, and 2 walks in 114 PA). If somebody reaches 2nd base against you with nobody out, that’s a couple more plate appearances with a man in scoring position than if that happens with 2 outs. It is true that the OBP against him with 1 out, .286, is his worst in the three “out” scenarios.

Reason #2: Wheeler hasn’t allowed a stolen base this year.

Here’s a point for you — the “leading off the inning” category is very similar to the 0-out breakout, particularly when you have the devastating success Wheeler has had there. But the two are not the same, as 0-out data keep compiling until you do, in fact, get an out. But Wheeler has limited leadoff batters to 17 times on base in 95 PA (.179), and batters hitting with 0 outs to 19 times on base in 114 PA. So it appears he’s only allowed the first two batters in an inning to reach base twice in 95 innings. (He’s faced 95 leadoff batters, but pitched 93 innings. So I would take it that the difference just represents the six outs from innings he has started but not gotten.)

6/26 N1: With exactly 1 home run per every 15 at-bats right now, Juan Soto is now backing up what was his career-best HR/AB last season (excluding the COVID season) with his second best. However, while he was with Washington, his BAbip was .314, and it’s .285 since. Neither 2023, 2024, or 2025 (so far) has been a .300 BAbip season for him.

I see a great player whose stats have shifted a bit from average to home runs, for whatever reason. One sometimes hears that he hits too many ground balls (and he certainly doesn’t hit too many fly balls), but it’s interesting to note that his three highest GB/FB ratios (this time including 2020) all came with Washington, when his BAbip was best. The constant with Soto, of course, is his incredibly high walk ratio, rather guaranteeing an outstanding rating for him, if he can just be good otherwise.

I was surprised to find that, for all his shifting of teams, over half of his career plate appearances have still been taken with Washington.

6/26 N2: Here’s something weird. Sacrifice flies in batting average constitute mulligans, but in BAbip, they count as outs. I think that makes for a needless additional variable when we compare the measures for a particular player. The right policy in one case should be the right one in the other.

6/26 N3: The impressive extent of the Royals’ ability to shut down an opponent’s running game was underscored on Wednesday when the base-stealing Rays singled 11 times against them but were not able to pick up a steal, with a caught stealing of Jake Mangum also in the equation.. This game also seemed to underscore the comeback of Salvador Perez as an excellent thrower. Perez has only given up 6 steals this year, while throwing out 8 runners.

The Royals allowed by far the fewest steals and attempts per opportunity last year, and are in front this year again in another cakewalk. Using data on Baseball Reference, opposing base stealers have only gone in 2.7% of the possible chances against them, with the Mets holding runners next best, at 4.9%.

Ovewr 2023-2024, Freddy Fermin was by far the better Royal catcher in throwing out runners, with a caught stealing rate of 37.1%, while Perez’s rate was less than half that, 18.0%. So you see Perez’s revival (remembering his earlier Gold Gloves). This year, Fermin’s been going along at his normal clip, with 10 steals against him in 15 attempts.

The two have very neatly split time over the three years they’ve been together. Perez has caught more innings in each year, but with percentages of only 60%, 54%, and 52%. But what is really interesting to me is that, even while the numbers said Perez was much easier to steal against, runners did not go against him more than they went against Fermin.

One thousand innings would represent an approximate workload for a starting catcher. From 2023-2025, Perez is seeing 67.2 attempts per 1000 innings, and Fermin is actually seeing a little bit higher 69.8 attempts per 1000 innings.

What that suggests to me is that the main reason the Royals see so few stolen base attempts isn’t their catchers, but their pitchers. The identity of the catcher hasn’t made a difference, despite the different success of the guys. The pushback would be that Perez has been a deterrent still on reputation, but given all of the miles he has on him, I think that’s a stretch, and think teams are smarter than that.

I do see 25 wild pitches allowed for Perez this year and 6 for Fermin, something relevant to their complete defensive performance. I assume they’ve been paired unevenly with different pitchers, and that could be the whole factor in that, but it’s something that stands out at a glance.

6/27 N1: I was prepared to make a whole sociological thing about Ben Rice somehow generating enough awareness that he has made the finals of the fan All-Star voting at DH (he will face off with Ryan O’Hearn). At least in the old days, if you didn’t have a name before the season, you would have to be great to enter public consciousness, and Rice hasn’t quite been that. But, looking closely, my narrative doesn’t quite pan out.

First, Rice made it, but this is the horse race where O’Hearn won by 12 lengths, and Rice just beat the rest of the stragglers coming home. It’s 1,762,125 votes for O’Hearn, and 674,120 for Rice. #9 Mike Tauchman still has 329,481 votes.

Then Al DHes have been less than stellar as a group, with a .715 OPS that ranks them 5th among the 9 positions (their .230 average is actually last). The weakness is underscored in the OPS for NL DHes being 2nd best among positions. Pre-season favorites Yordan Alvarez and Anthony Santander have missed time or been bad, or both.

I still would have expected many more votes for Yandy Diaz (354,175), and some more for Jorge Polanco (338,878), with what Polanco did early in the season. He was platooning, but wasn’t far off from Raleigh.

6/27 N2: As his career rate is 0.86, it’s not a huge difference, but I do note that Matt Olson’s GB/FB rate of 1.06 this year is the highest of his career. He’s been playing 10 years.

6/27 N3: It seems we usually hear about Ketel Marte’s position as one of the best players in the game in the context of his complete skill set, but at this point in his career, the man flat out mauls the baseball. There is nothing subtle about the danger he presents. Even with all of the time he has missed over the last two years (he is 99th in plate appearances), he is 10th in home runs. If we take players with 800 plate appearances, he has the 4th best home run-per-plate appearance rate, with only Judge, Ohtani, and Raleigh ahead of him. If we drop the plate appearance requirement to 400, he is still 4th.

A reminder of where he has come frm…..Had just 3 home runs to start his career in 2015-2016, in 177 games….Through 2023, he still had just 107 career home runs in 968 games, with 968 games 4 shy of 6 162-game seasons….

6/28 N1: Although they lead the National League in E.R.A. (and with only the Giants in close pursuit, I won’t say they “still lead”), the Mets lead in walks, too. They led the NL in walks last year, but placed just 7th in E.R.A.

While the team’s pitching has not been good in June, the control leak started in May when they issued 4.28 walks-per-9 yet still had a 3.15 E.R.A.. This month, they’ve walked 3.81-per-9 but have a 4.30 E.R.A.

Since in May, they gave up 24 home runs in 27 games and this month they’ve given up a similar 22 in 25 games, I’m not quite sure why their performance has fallen off so decidedly. I wondered if they’ve been poor when teams have had runners in scoring position against them, but the batting average then has been just .249.

In getting that list, I noted a couple of extremes: first a .145 allowed in June for Houston with runners in scoring position; then, a .346 allowed for Minnesota. Pittsburgh has the second-lowest RISP average allowed in June, at .199, and Atlanta has the second worst, at .299.

The Mets’ pitching staff is marked by the almost universal lack of stinginess when it comes to walks. No one drives you crazy walking batters, but walks are a constant part of the experience.

They have 11 pitchers with 20 or more innings. The possibly-done-for-the-season Max Kranick has only allowed 5 walks in 37 innings, or 1.2-per-9. But the second best rate of these 11 pitchers is David Peterson, whose mark is an average 3.2-per-9. Of course, the Mets would love to just have their initial starting staff still intact, walking batters or not, but Baseball Reference shows the Mets starting five of Peterson, Holmes, Canning, Senga, and Megill with 34, 35, 35, 31, and 33 walks. From this, you’d think the Mets were throwing from 80 feet 6 inches, while the other teams were throwing from the standard 60’ 6”.

The Angels are also repeating as “league walk champion” this season. The White Sox, fourth overall in MLB in walks last year, are second this year. So there happens to have been a lot of commonality in the statistics. The A’s are the new team that has snuck in the MLB top three, after being tied for 8th last year.

6/28 N2: We’re halfway through the season and, buoyed by the 6 runs Jose Caballero gave up, the Orioles just beat the Rays 22-8. So it seems a good time to report on position-player pitching this season. And, unfortunately, there is news to report. The use is greatly increased from a year ago, and the offensive party has been all the more uproarious. Consider

Last year, hitters hit against position players 445 times. This year, the total is already at 405. Doubling the number, that’s an 82% increase.

Probably most alarming — last year, position players gave up 80 runs. This year, they’ve already given up 147!

As a point of fact, that raises the possibility that batters faced are just up because the position players can’t get anybody out, so the merry-go-round goes round and round. But innings, too, are on pace to be up 58%.

As offense hasn’t seen a real increase this year compared to last (at least when position players aren’t pitching!), it doesn’t make much sense that there would be an increase in lopsided games that has greatly expanded their use. It seems more likely that teams are really seizing every opportunity to use position-player pitchers, and that the whole thing has just become completely lax, both in terms of any misgivings that this constitutes an embarrassment, and any sense that the position-player pitchers should be doing anything but supporting the fraternity of ballplayers on the other team with some freebees.

6/28 N3: No one in the Red Sox Friday lineup (Duran, Toro, Anthony, Story, Abreu, Rafaela, Hamilton, Wong, and Sogard) exited the game with better than a .324 OBP for the season. I give a free pass on such things when the other team starts a left-hander, but this was against Jose Berrios, who perhaps not entirely surprisingly threw 7 shutout innings.

6/28 N4: The one thing with the position-player pitching and the lobbing of the ball, like last year, the home runs have been very much in control. Just 19 allowed so far.