Question for scea about clutch

[tessl]

Would you please explain exactly how clutch works, when it is activated and what impact it has.

Thank you.

[swaldo]

Re: Question for scea about clutch

Just for info, here is the links to the tests I have conducted regarding clutch ratings:

http://www.operationsports.com/forums/scea-sports-mlb/313311-comeback-ai-thread-7.html#post2039472238
<O</O
http://www.operationsports.com/forums/scea-sports-mlb/313311-comeback-ai-thread-8.html#post2039476013
<O</O<O</O

[moemoe24]

Re: Question for scea about clutch

Just guessing here, but I would think that clutch kicks in situations like....men on 2nd and 3rd and 2 outs. A player with a higher clutch rating would be more likely to get a hit.

Another example may be your closer coming in with a man in scoring position. If his clutch is high he will get out of the jam easier than a pitcher with a low clutch rating.

I know first hand that clutch effects your closers. I use Lindstrom from the Marlins. He blew 3 saves in 3 games. I thought that was weird so I checked his ratings. Sure enough, his clutch rating was in the 60's. I moved it up to 80 and he hasn't blown one since.

That's my thought on clutch anyway. It may be something totally different.

[Revolution42]

Re: Question for scea about clutch

I would think that your analogy is accurate....situations occur when a clutch performance is required, player with the higher clutch attribute rating will have a better chance of making the big play....key hit....perfect pitch etc etc...sounds about right I would think.

[tessl]

Re: Question for scea about clutch

Just guessing here, but I would think that clutch kicks in situations like....men on 2nd and 3rd and 2 outs. A player with a higher clutch rating would be more likely to get a hit.

Another example may be your closer coming in with a man in scoring position. If his clutch is high he will get out of the jam easier than a pitcher with a low clutch rating.

I know first hand that clutch effects your closers. I use Lindstrom from the Marlins. He blew 3 saves in 3 games. I thought that was weird so I checked his ratings. Sure enough, his clutch rating was in the 60's. I moved it up to 80 and he hasn't blown one since.

That's my thought on clutch anyway. It may be something totally different.

I've got a general idea of what it is but I'd like to know the exact details so I know how the game works. I suspect is plays a far greater role than many people are aware of and is distorting the statistics.

One of the biggest home runs in baseball history was hit by Bill Mazeroski and he was never really a "clutch" hitter. Never hit 20 home runs in a year in 17 MLB seasons, never had 100 rbi. On the other hand one of the great clutch hitters, McCovey made an out with the world series on the line.

I'd like to get rid of it completely but short of that I'd like to know exactly how it works.

[Millennium]

Re: Question for scea about clutch

I've got a general idea of what it is but I'd like to know the exact details so I know how the game works. I suspect is plays a far greater role than many people are aware of and is distorting the statistics.

One of the biggest home runs in baseball history was hit by Bill Mazeroski and he was never really a "clutch" hitter. Never hit 20 home runs in a year in 17 MLB seasons, never had 100 rbi. On the other hand one of the great clutch hitters, McCovey made an out with the world series on the line.

I'd like to get rid of it completely but short of that I'd like to know exactly how it works.

Get rid of it completely? So, you don't want any difference between Jeter and Robinson Cano in the bottom of the 9th, 2 outs and the winning run on 2nd?

[Melkro]

Re: Question for scea about clutch

Get rid of it completely? So, you don't want any difference between Jeter and Robinson Cano in the bottom of the 9th, 2 outs and the winning run on 2nd?

How many times does it have to be said...clutch hitting is an illusion. It has been statistically proven as such many times. Jeter (for as "clutch" as people claim he is) has a lower career average in the playoffs (the ultimate clutch situation) then he does in his regular season career. he is preceived as being a good "clutch" performer because he is a good performer in all situations. In Jeter's case the illusion is only made worse by the fact that he plays in New York AND has been to the playoffs enough that random chance predicts he should have several key hits in his postseason career. To answer your question yes I would certainly take Jeter over Cano in that situation, but only because Jeter is a better hitter than Cano in all situations.

I'm with the OP in that I think clutch needs to be taken out of the game. There is no proof beyond aligory that it exists and stats show it doesn't, so why does the most realistic baseball game of all time include it? At the very least allowing users to turn it off would be nice.

http://www.geocities.com/[email protected]/CramerClutch2.htm

http://www.baseball1.com/bb-data/gra...ullclutch.html

"The correlation between past and current clutch performance is .01, with
a standard deviation of .07. In other words, there isn't a significant
ability in clutch hitting; if there were, the same players would be good
clutch hitters every year."


Start your reading there

[SoxFan01605]

Re: Question for scea about clutch

^^^how did I know the Bill James disciples would come out in full force on this.

Jokes aside, ratings like "clutch" can be good for videogames because they allow for a bit of variance in performance. Meaning a good clutch rating for a lower skilled player can give him a better chance of performing in bigger situations if he's been known historically to come up a bit bigger than his skill level suggests normally.

The trouble with the notion of zero clutch (in a videogame), is that it tries to paint baseball as a black and white game of statistical definites. That is fine for analyzing the real game, because it's impossible to truly eliminate the psychological and basic human element from the game in real life (it's always there, despite whatever statistical model you base your analysis on). So analyzing what you can purely on a solid statistical base makes pefect sense and should be better embraced.

In a videogame, you are starting from black and white and essentially have to program variance or the "human" element. It's obviously never going to be perfect and will have it's own set of hiccups, but I'd rather have an imperfect attempt at inperfection (which only enhances realism if done well) over a model free from any deviation whatsoever.

I would agree with the notion it potentially needs to be tweaked or toned down perhaps (depending on just what exactly it does. Some assume more simply because of mere verbiage...lol...much like the great "comeback AI" beliefs of some).

Clutch itself isn't a myth (even the most diehard disciples should know that), the "myth" is the notion that a given player is more "clutch" than his skill level would suggest in certain situations over a large sample.

Also, there are things statistic can't yet fully evaluate because we don't know enough to isolate certain variables (Stress level, peak, mental state, etc.). These could very well support the "myth" that some players are better tuned to perform above their peak in given situations. It would still be more anamoly than certainty (meaning it would not be as simple as player "A" is more/less clutch than "B". Over the long haul skill would still be more of the determining factor), but there is still much unknown in the debate...despite what some might tell you.

[Revolution42]

Re: Question for scea about clutch

^^^how did I know the Bill James disciples would come out in full force on this.

Jokes aside, ratings like "clutch" can be good for videogames because they allow for a bit of variance in performance. Meaning a good clutch rating for a lower skilled player can give him a better chance of performing in bigger situations if he's been known historically to come up a bit bigger than his skill level suggests normally.

The trouble with the notion of zero clutch (in a videogame), is that it tries to paint baseball as a black and white game of statistical definites. That is fine for analyzing the real game, because it's impossible to truly eliminate the psychological and basic human element from the game in real life (it's always there, despite whatever statistical model you base your analysis on). So analyzing what you can purely on a solid statistical base makes pefect sense and should be better embraced.

In a videogame, you are starting from black and white and essentially have to program variance or the "human" element. It's obviously never going to be perfect and will have it's own set of hiccups, but I'd rather have an imperfect attempt at inperfection (which only enhances realism if done well) over a model free from any deviation whatsoever.

I would agree with the notion it potentially needs to be tweaked or toned down perhaps (depending on just what exactly it does. Some assume more simply because of mere verbiage...lol...much like the great "comeback AI" beliefs of some).

Clutch itself isn't a myth (even the most diehard disciples should know that), the "myth" is the notion that a given player is more "clutch" than his skill level would suggest in certain situations over a large sample.

Also, there are things statistic can't yet fully evaluate because we don't know enough to isolate certain variables (Stress level, peak, mental state, etc.). These could very well support the "myth" that some players are better tuned to perform above their peak in given situations. It would still be more anamoly than certainty (meaning it would not be as simple as player "A" is more/less clutch than "B". Over the long haul skill would still be more of the determining factor), but there is still much unknown in the debate...despite what some might tell you.

Excellent post.....thanks.

[Phoenixmgs]

Re: Question for scea about clutch

I definitely think some players are clutch. Some players can just focus in more in key spots or maybe just go up to the plate with a better strategy compared to at bats with no one on base. As a Cubs fan, I remember Ricky Gutierrez (and yes, I said Ricky Gutierrez of all people) being super clutch in 2001. I just remember him always putting together a solid AB in big spots, he hit a grand slam to beat the White Sox in that season and I'll never forget that HR. So, I looked up his situation stats that year and here they are:

From http://ca.sports.yahoo.com/mlb/playe...1&type=Batting

Overall: AVG .290 - OBP .345

Bases Empty>>> ABs 311 - AVG .280 - OBP .327
Runners On>>>> ABs 217 - AVG. 304 - OBP .367
RISP>>>>>>>>> ABs 130 - AVG .293 - OBP .368
RISP w/2 outs>> ABs 77 - AVG .273 - OBP .385
Bases loaded>>> ABs 16 - AVG .313 - OBP .300

None on/out>>>>>>>> ABs 111 - AVG .297 - OBP .345
None on: 1/2 out>>>> ABs 200 - AVG .270 - OBP .318
Men on, 2 outs>>>>>> ABs 105 - AVG .267 - OBP .364
Man on 3rd, <2 outs>> ABs 23 - AVG .348 - OBP .257


I think these stats definitely back up the fact that Ricky Gutierrez gave the team better at bats in important situations. His bases empty AVG is less than his average with runners on base and his OBP is much higher when men are on base (except bases loaded but he only had 16 ABs, very small sample size). Ricky Gutierrez basically knew how to bat based on the situation; even looking at his 'Man on 3rd, <2 outs' stats, you can see he went up with the goal of putting the ball in play to get that runner home as his OBP in that situation is much lower than in any other situation and his AVG is nearly 100 pts higher than his OBP.

[SoxFan01605]

Re: Question for scea about clutch

^^^ The trouble with these stats is that they prove little in the debate of whether or not players are clutch. The sample is too small. The argument against it is that over time those numbers will level off. Meaning a player will perform to the level of his ability, no matter the situation, over the long haul. I'm not saying your stats are wrong, just that they are inconclusive in the debate of a clutch player. They merely prove he was "clutch" over a period of time.

I somewhat agree with this line of thinking, but only in the most general sense. Those that have bought into the newer line of thinking on these topics tend to blanket such views and distort the point of related theories a bit.

I'm not in full compliance will the disciples though...lol...I believe some players are better wired (in psychological make-up) to perform in high stress situations. I completely disagree with the notion that all MLB players have the make-up for this (which is a foundational premise for the argument against "clutch") simply because they made it to the top league. EVERYONE has a peak.

I understand the argument against "clutch", but the trouble is it places too heavy a value on ability and almost completely ignores make-up. I would say it's driven by sound principles but unfortunately, like most new theories, has been bought in to with too much zeal to see it's holes.

Again though, I think this is a sidebar to whether or not it should be in the game. I think it has more weighted value when trying to add variance to rigid parameters than it has when trying to analyze the real life game.

BTW, on a somewhat loosely related topic, this Yankees/Red Sox series is killing me thus far...lol. There's so much "clutch" and "comeback" in these first two games that it's giving me heartburn. It's been some serious back and forth craziness thus far.

[Melkro]

Re: Question for scea about clutch

^^^ The trouble with these stats is that they prove little in the debate of whether or not players are clutch. The sample is too small. The argument against it is that over time those numbers will level off. Meaning a player will perform to the level of his ability, no matter the situation, over the long haul. I'm not saying your stats are wrong, just that they are inconclusive in the debate of a clutch player. They merely prove he was "clutch" over a period of time.

I somewhat agree with this line of thinking, but only in the most general sense. Those that have bought into the newer line of thinking on these topics tend to blanket such views and distort the point of related theories a bit.

I'm not in full compliance will the disciples though...lol...I believe some players are better wired (in psychological make-up) to perform in high stress situations. I completely disagree with the notion that all MLB players have the make-up for this (which is a foundational premise for the argument against "clutch") simply because they made it to the top league. EVERYONE has a peak.

I understand the argument against "clutch", but the trouble is it places too heavy a value on ability and almost completely ignores make-up. I would say it's driven by sound principles but unfortunately, like most new theories, has been bought in to with too much zeal to see it's holes.

Again though, I think this is a sidebar to whether or not it should be in the game. I think it has more weighted value when trying to add variance to rigid parameters than it has when trying to analyze the real life game.

BTW, on a somewhat loosely related topic, this Yankees/Red Sox series is killing me thus far...lol. There's so much "clutch" and "comeback" in these first two games that it's giving me heartburn. It's been some serious back and forth craziness thus far.

I understand the arguement you make, but the truth is if some players had a psychological makeup that enabled them to perform better in clutch situations then stats would show that. I'm not saying there are no clutch performances, because there obviously are, it's just that players themsleves are not clutch. If anyone can find an example of a player that consistently played above his career stats in clutch situations I'll certainly retract my statment, but it just doesn't happen. Players have clutch "streaks"...just like the guy that was talking about Gutierrez said. But the truth is that is bound to happen from time to time...if Gutierrez truely had some special psychological makeup that made him perform better in clutch situations his whole career should show this trend...which it does not.

I do understand the arguement that clutch ratings give videogames more randomness, but it's flawed logic. If randomness (and the possibility that a crappy player might come up big) is what you are looking for that possibility should be applied equally across all players, not given to a select few that the general public mistakenly believe are "clutch players." Secondly, the top clutch ratings in video games almost exclusivly go to the best players who people think are clutch because they come up big in clutch situations more often then crappy players...which is simply logical because they are better players. So the best players in the game end up with an unwarrented boost in key sitations.

It seems that this could be handled better in video games to me. If a bit of randomness is what people want (I know I like it in my sports games) why not make it truely random rather then based on some completely unfounded rating?

[davewins]

Re: Question for scea about clutch

I would like to see no clutch rating in the future at all. This is if it really gives bogus attribute bonuses during a player's at bat. I would rather just have clutch determined by me controlling the game. I would like to just have to concentrate that much harder in a clutch situation to get a hit rather then have normal concentration at the plate and have some attribute bonus because that player has a high clutch rating.

[moemoe24]

Re: Question for scea about clutch

How many times does it have to be said...clutch hitting is an illusion. It has been statistically proven as such many times. Jeter (for as "clutch" as people claim he is) has a lower career average in the playoffs (the ultimate clutch situation) then he does in his regular season career. he is preceived as being a good "clutch" performer because he is a good performer in all situations. In Jeter's case the illusion is only made worse by the fact that he plays in New York AND has been to the playoffs enough that random chance predicts he should have several key hits in his postseason career. To answer your question yes I would certainly take Jeter over Cano in that situation, but only because Jeter is a better hitter than Cano in all situations.

I'm with the OP in that I think clutch needs to be taken out of the game. There is no proof beyond aligory that it exists and stats show it doesn't, so why does the most realistic baseball game of all time include it? At the very least allowing users to turn it off would be nice.

http://www.geocities.com/[email protected]/CramerClutch2.htm

http://www.baseball1.com/bb-data/gra...ullclutch.html

"The correlation between past and current clutch performance is .01, with
a standard deviation of .07. In other words, there isn't a significant
ability in clutch hitting; if there were, the same players would be good
clutch hitters every year."


Start your reading there

Thats BS. I'm not saying you are full of BS, but the baseball minds can try to convince me clutch hitting is an illusion all they want....it's just not true. What do you think a good average with RISP means?? It means the guy who's batting .330 with RISP is more clutch than the guy batting .215 with RISP. Maybe Jeter is a poor example of a clutch hitter, or maybe not, I dont know, but I do know that there are some guys that seems to get a hit when their team needs one. I dont know how that can be proven as an illusion.

[SoxFan01605]

Re: Question for scea about clutch

lol...this is what I mean about people living by a gospel. I'm not even sure if you completely read what I wrote. You're almost exclusively quoting Bill James and the like. That's fine, of course...it's your right. I just don't 100% buy into all the premises of the theory as some do, is all. No biggie...I probably shouldn't have even jumped into it because I knew it would get sidetracked from the point about the game...lol. Oh well.

That said, I do agree about clutch streaks...that's actually what I alluded to. I don't believe I once said there are "clutch" players. My point was there are players who have a better make-up to not buckle in certain situations, regardless of physical skill...it's the same in reverse. It's hard to measure statistically because there are too many variables in play. I also said it would be situational (agreeing that there are no specifically "clutch" players) only be a factor and not a determining factor. Meaning skill is still the predominant and overriding factor in success. Let's not confuse this with inferring I suggested "magical" psycholgical powers...lol.

Jokes aside, I think some people blanket the argument on both sides. "Clutch" is the most overargued and misinterpreted terminology in baseball, IMO. People get all tangled up in the word rather than what it means in context.

Also, I agree that the randomness could be handled better. I'm guessing (being nothing close to what even resembles a programmer...lol) that it's an on-going process. People have felt the AI "cheated" in sports games for years. I'm sure at some point, the fact that you're playing through programmed events will feel as such.

My point was (and I feel it's being beared out) is people overreact to terminology. People have done the same thing with the "CPU Comeback" option. There infer "comback code" or "rubberband AI" simply because there's an option with the term "comeback" in it. I'm not talking about the debate of whether or not there is something screwy with the AI when it's down...simply the argument that it must be there since the option is there. This is what I mean about being hung up on terms.

"Clutch" is just a term. Even if it was some other random variable under the hood determining performance trends, etc...it would still be programmed. As I said, I think people overreact to the word "clutch."

I think if SCEA would expound on what it means in terms of how the game plays, then it might settle things down a bit (remember, that was the real point of the thread which we all strayed from). I'd be curious to hear it as well. I'm just not spending my time worring about it and running constant tests over enjoying the game. More power to those who do, of course. I respect the effort and thoroughness. I'm sure their work could end up rewarding to them in a sense...it's just not for me.

The fact is, I personally don't care what they call it or how they implement it...as long as it bares out realistically in the game. That was my point, "clutch" can be useful to the videogame representation if it's part of a system that works well. SCEA's game does an outstanding job of representing the real game, most would agree. I'm sure there are things they can tweak and improve (maybe "clutch" or related attributes, like confidence, fall into that), but I don't believe a total removal is necessary.