I guess clutch is highly overrated

A different view on this discussion. Happened to stumble across this other team of mine. That is "similar" in terms of games played and offensive production, not equal certainly, but similar enough that I wouldn't have expected this disparity.

Why the HUGE disparity?

84 Games and 2 Clutch Opportunities = 1 Clutch Opp every 42 games:
https://365.strat-o-matic.com/team/1759040
https://365.strat-o-matic.com/team/misc/1759040

vs.

102 games and 34 Clutch Opportunities = 1 Clutch Opp ever 3 games:
https://365.strat-o-matic.com/team/1757684
https://365.strat-o-matic.com/team/misc/1757684

Yes, the first team has a better offense, but it's not that much better. Yet it has 14 times the number of clutch opps in a similar number of games.

And I'm not saying there's anything nefarious going on. It's just odd that there is such a large difference.

One thing I noticed is that the team with more chances happens to lead the league in both Doubles AND Triples, but is that really enough to explain that big of a difference?

Or is it simply the blind squirrel finding a nut more often type of situation???

And I'm not saying there's anything nefarious going on. It's just odd that there is such a large difference.

One thing I noticed is that the team with more chances happens to lead the league in both Doubles AND Triples, but is that really enough to explain that big of a difference?

Or is it simply the blind squirrel finding a nut more often type of situation???

One factor I notice is that the team with two clutch opportunities only has a 0.342 OBP, whereas the other team has a 0.378. Those kind of fall on the opposite ends of the spectrum as 0.342 is not very good, but 0.378 is decent. That means the one team is getting OB 13.2 times per game while the other is 15.5. Perhaps that extra 2.3 OB per game is pushing more runners into scoring position leading to more clutch opportunities. My guess is that is part of the difference and good ole statistical luck is the other part.

I had a pretty good hitting team finish the other day. They had a 0.384 OBP and 15.6 OB per game. They ended up 0-25 in clutch, so one chance every 6.5 games.

In my NERP formulas, I scored 1 Clutch hit per side (2x weight for vs R, 1x weight for vs L) as 0.16 NERP. That seems way to high, which is fairly predictable.

I don't have my notes in front of me of why that much. 1 "SI" on a diamonddope card (so SI in the #2 row, 1 outcome per 108 PAs) scores 0.568 NERP. Giving about 28% (0.16 NERP) as much for a clutch SI seems too much.

Here are basic numbers. Can't we figure out the % of PAs that would occur in "clutch" situation? ---> 2 outs with runner(s) in scoring position? Chat GPT roughly states "historically, around 20-25% of plate appearances in Major League Baseball occur with two outs and one or more runners in scoring position." -------> this is laughably high.

BBref has League Splits by year. https://www.baseball-reference.com/leag ... &year=1996

% of PAs that occured with 2 outs, RISP
From 2014-2013: 11.45% --- (all PAs = 0.250/0.319/0.412)
From 1996-2001: 12.50% --- (all PAs = 0.268/0.339/0.427

Interestingly, every year in the 16-year sample shows "clutch" OPS being about 15 to 20 points lower. ONLY 2021 shows better #s in 2 outs and its by a tiny amount. 2021 RISP: 0.712 vs 0.707 overall.

Going from +16 clutch to -16 clutch can be calculated as a percentage of those 108 PAs on one side of a hitters card. This is about the whole range of clutch hitters.
+/-16 clutch = 14.8%
+/-15 clutch = 13.9%
+/-14 clutch = 13.0%
+/-13 clutch = 12.0%
+/-12 clutch = 11.1%
+/-11 clutch = 10.2%
+/-10 clutch = 9.3%
+/- 9 clutch = 8.3%
+/- 8 clutch = 7.4%
+/- 7 clutch = 6.5%
+/- 6 clutch = 5.6%
+/- 5 clutch = 4.6%
+/- 4 clutch = 3.7%
+/- 3 clutch = 2.8%
+/- 2 clutch = 1.9%
+/- 1 clutch = 0.9%
+/- 0 clutch = 0.0%

Crossing that with hypothetical %s of "Clutch" PAs, regardless of the batters "clutch" symbols, gives us a matrix of %s for each clutch rating by ~"Clutch PA %"

Here's a basic outline for a few clutch ratings. This is % of PAs on the HITTER's card that should be clutch symbols. It's just the % of the card that is clutch symbols x the percentage of all PAs that should be "clutch" situations. Yes, half this many PAs would be "clutch" eligible because the pitcher cards (1/2 of PAs) don't have clutch symbols. Still for finding NERP, we're looking at just on the hitters' 108 PAs. Right?

+/- 10 clutch @ 11% Clutch PAs = 1.02%
+/- 10 clutch @ 12% Clutch PAs = 1.11%
+/- 10 clutch @ 13% Clutch PAs = 1.20% --- This is 1.29 SI on the card for 10clutch, which is 0.07 NERP per $ symbol?
+/- 10 clutch @ 14% Clutch PAs = 1.30%
+/- 10 clutch @ 24% Clutch PAs = 2.22%
+/- 10 clutch @ 25% Clutch PAs = 2.31%
+/- 10 clutch @ 26% Clutch PAs = 2.41%
+/- 10 clutch @ 27% Clutch PAs = 2.50%

I think this is right? Originally I was looking at 20%+ "clutch" situations, but 11-12% has been the MLB range. So, what's the % of PAs in ATG that occur with 2 outs and runner in scoring position? I think it'd be higher with the high OBPs, XBH, runs and SB #s in many league? 12.7% in MLB 1999 and 2000 when league OPS was .778 and .782 respectively ---- the highest OPSs ever. So 13-14%? That shows 0.072 NERP per $ symbol if they're valued exactly as a single. That's less than half of my previous 0.16 NERP value. Some "contextual" benefit (clutch hits = runs scored, so wins or at least more RBI stats which we love stats!) So what % of a normal SI should a clutch SI be worth? 1.5x? That'd get it to 0.108 NERP which is closer (2/3rds) of my original 0.16 NERP? ----- Will adjust I think ---

nels's numerical work is very impressive indeed. Meanwhile, I decided to look at the numbers from a different angle to check the validity of my claim that a manager can safely ignore clutch if they don't want to bust their brains, especially since most good hitters have neutral or negative clutch. I recently played in a 24-team Barnstormer 23 Semi-final #2 league. I figured a 24 team league would generate enough results to be meaningful. Anyone can find the stats for this league or any Barnstormer league here: https://365.strat-o-matic.com/tournament/47 To tease out the clutch results will require further digging.

For starters, I looked at the playoff teams, which I figured were the most successful. How much effort did the managers of these successful teams seem to be putting into finding clutch hitters? And how did the teams make out? I've indicated in parens the playoff teams that won 90 or more games, as well as indicating my own team.
Playoff teams: Clutch H - O
0-23 Bresee (95 w)
2-15 Brownstone (91 w)
2-16 Asgardian (93 w)
0-24 Charleston (91 w)
0-19 Celtic
1-23 Republic (my team; 87 w & a ring)
0-9 Mudville
6-35 La Rose (Appling 4-0 w/ 19 IBB; Anson 2-0 w/ 16 IBB)
Totals: 11 clutch hits - 114 outs

Then, having looked at the playoff teams, I thought I ought to look at clutch outcomes for the non-playoff teams a well. So, here goes. I've indicated IBB totals for positive clutch hitters where they seemed notable.

Non-Playoff Teams: Clutch H - O
6-10 Mojave (Lajoie: 3-0, 20 IBB)
9-15 Homestead (Izturis 2-0, 21 IBB; Gardner 5-0, 17 IBB; Alomar 2-0 16 IBB. 83-79 w/l)
4-19 Adelphi
0-12 Sunset
1-11 Bingo
14-14 Hardtack (73 w)
1-32 Copley
5-15 Southlake
0-42 Amish
3-25 Wayne
3-26 B-Town
3-29 Brightwood
3-18 On DL
1-16 BS
2-25 Bartering
1-8 Stormin
Totals: 51 clutch hits - 327 outs

Now lets add up the playoff and non-playoff teams. What are the clutch totals for the whole 24 team league?
League Totals: 62 clutch hits - 471 outs
The league generated 135817 at bats. So without attempting to account for IBB, clutch hit & out events amounted to 0.0039 of all AB. Expressed in percentage terms, clutch hits & outs amounted just under 4/10 of 1% of all AB. Clutch hits accounted for 62 of those 135817 AB. There were 36640 hits in the league. 62 of these were clutch. That's 0.0017 of all hits in this 24-team league. Six teams had zero clutch hits, one team going 0-42.
I'll make some observations in a separate reply.

Observations about Clutch based on the results of Barnstormer 23 Semi-final #2 league.

*Clutch hits are very rare events, and some teams will have no clutch hits, or very few, for an entire season.
•There is little or no correlation between team clutch hits vs outs and overall team performance. The team with the most wins in the league (95) had 0-23 clutch. The team that emerged as champs were 1-23.
•FrankieT is absolutely right that positive clutch has its most important impact on IBB. Players who achieved clutch hits who appear in the top 12 leaders in IBB include Appling, Anson, Izturis, Gardner, & Alomar. These players totaled just 16 clutch hits but 89 IBB. That's about 5 IBB for every clutch hit. It's hard to track IBB beyond the list of 12 leaders without reading through all the box scores for a 24 team league, which I aint gonna do.
•Most hitters in ATG have negative or neutral clutch. Players with massive RBI totals are frequently penalized. When Sammy Sosa drove in 160 runs in 2001, his clutch ratings were -6 / -7. But whether they drive in a lot of runs or not, most players in ATG have negative clutch ratings. This is underlined by the 62-471 hit - out performance in Semifinal #2.
•Players gifted with positive clutch are often hitters with little or no power and mediocre batting average who hit a few more RBI than expected. Izturis is a good example. Some good deadball hitters, e.g. Appling, Anson, Lajoie, are also rewarded.
•If you want to build a power team, you more or less have to say goodbye to clutch. This is not necessarily a bad move.
•I believe clutch is priced in to player cost, so if you get a positive clutch for a player, you will be paying for it.
•You really have to get down in the weeds to find a clutch hitter who is also a bargain.
•Clutch as implemented by SOM makes no sense at all, assuming the goal is to generate RBIs for players with significant clutch ability. More often than not, these players are intentionally walked, because HAL over-reacts to the clutch rating. The handful of positive clutch hitters are therefore going to become disproportionately frequent baserunner, but they get fewer than expected RBI.
•I'm going to double down on my practice to set IBB as Extra Conservative for teams and individual pitchers. Why issue an IBB to Gardner or Izturis, and as a result likely face a better hitter with an extra runner on base?

By the way, here are the numbers for the top twelve Intentional Walk leaders in the Semifinal #2 league. The list includes player name, team abbreviation, number of IBB, and clutch rating. In general, this list confirms FrankieT's contention that HAL disproportionately IBBs players with positive clutch.

---INTENTIONAL WALKS
L.O'Doul COP 23 -6/-4
C.Izturis HOM 21 +3/3
N.Lajoie MOJ 20 +4/4
L.Appling LAR 19 +9/8
G.Goslin COP 17 +3/3
E.Howard 201 17 +2/2
L.Gardner HOM 17 +7/5
J.Foxx 201 16 +4/4
R.Alomar HOM 16 +3/2
C.Anson LAR 16 +5/5
G.Sisler CEL 15 0/0
P.Traynor BSD 15 +8/8
Total IBB = 212

Ten of these twelve IBB leaders have positive clutch--none have all that many clutch hits, the most being 5 by Gardner. In light of HAL's tendency to IBB positive clutch players, I can't explain why O'Doul, who has negative clutch, leads the league in IBB. A total of 985 IBB were issued in the league, so these 12 players, with 212 IBB, account for 22% of all IBB.

following this thread, fascinating, and the depth of research and analysis incredibly impressive

may i inquire as to the recent IBB aspects ?

although the clutch "hits" recorded weren't very high, isn't an IBB kind of a hit ?.....(Little League coach in my ear, "a walk is as good as hit")

it is an on base opportunity, and doesnt create an out

of course, like any BB, it isn't quite as good as a hit, but should those IBB be factored in somehow to the analysis ?

thanks

Some interesting insights to be considered here. My biggest takeaway is the positive clutch leads to Intentional Bases on Balls. Meaning you might be able to put a so-so hitter in the 5 or 6 hole who has a positive clutch and they'll end up having an elevated OBP because they might get IBB'd 20 times in a season. And if nothing else, I will now make sure that my big boppers have a negative clutch so that they will be less likely to get walked intentionally.

Potential Lineup Thoughts:

1- HIGH OBP - MAYBE HIGH IN DOUBLES AND/OR TRIPLES
2 - HIGH OBP - LOW IN gbA'S - ALSO MAYBE HIGH DOUBLES AND/OR TRIPLES
3 - HIGH OPS - BAD CLUTCH (TO AVOID GETTING WALKED INTENTIONALLY)
4 - HIGH OPS - BAD CLUTCH (TO AVOID GETTING WALKED INTENTIONALLY)
5 - CHEAP VALUE PLAYER WITH A POSITIVE CLUTCH
6 - CHEAP VALUE PLAYER WITH A POSITIVE CLUTCH
7 - 4 - HIGH OPS - BAD CLUTCH (TO AVOID GETTING WALKED INTENTIONALLY)
8 - CHEAP VALUE PLAYER WITH A POSITIVE CLUTCH

Crazy stuff with IBB. Likely unconsidered Still, I can't think it changes the value of clutch that much, right? Clutch is still "good,"----OuttaLeftfield shows the high-quality play of the Barnstormers and how it lacks positive clutch hitters. This doesn't mean clutch is BAD, just that the best teams and/or best hitters didn't happen to use it. $100 mil leans towards those MONSTER hitter more etc. Lot to digest with whether "negative" clutch Elite Hitters are better because they're pitched to more.... Too bad we can't see IBB #s except for the top 12 in the league via leaders?

But after looking at just one other league, the Cesar Izturiss of the world do make appearances in the "Leaders" for IBB. $1mil Catcher Mike Tresh gets 17 IBB behind a .207/.315/.235 line. His BR is -0.68/-2.73 (yikes). With his +8/+8 clutch as pure hits on DiamondDope, he's about +5/+3 BR. Terrible, yet worthy of tons of IBB. He only has 0.5 and 0.6 2B on each side of the card---no other XBH.

Very big stuff uncovered OuttaLeftfield. I really stopped looking at the Leaders Page I guess. But yeah, Lajoie/Appling are common names there----but apparently so are terrible hitting "good clutch" guys.

Here is data from seven $80mil sims in ATG8 from back in ~2016
-13 of the 77 "League Leaders" in IBB were under $2mil dollars. All had positive clutch and averaged about +7.5/+7.5 clutch symbols and 16 IBB. ---- these are mostly terrible hitters mind you
-All 77 League Leaders averaged +3.8/+3.6 clutch and 17 IBB.
-8 of the 77 League Leaders have "0" clutch and 15 have "negative" clutch. This is about 1/3rd of all leaders being NON-Clutch hitters.
-The names of those 13 sub-$2mil IBB stars: Joe Sewell, C. Grimm, B. Boone x2, J. Kling x4, C. Travis x2, M. Tresh, H. Zimmerman, B. Richardson.
-Players can have 30-50% of their Walks on the year come from IBB. Gotta wonder how the playability of a cheapo like Bobby Richardson depends on if he's getting those 15+ extra walks from a IBB from HAL....
-Charlie Jamison at $2.37 got 12 IBB for a 74-win team with .371 team OBP. Jamieson has 0/0 clutch so he's the lone "bad bat" without Clutch to be IBB'd alot. *This isn't that much IBB at 12, so maybe more of these exist just outside a league's "top 12"