enhancement requests for Bernie

[MARCPELLETIER]

[quote:1bcb1c4897]We're getting off topic a bit MW, but I disagree with that. I think if SOM is correctly replicating the OB & baserunning of the teammates in its replay system, and the opponents, environment, etc., then SOM should be able to replicate RBI results accurately without adding an artificial RBI adjuster.[/quote:1bcb1c4897]

Yeah, getting off topic, so this will be my only post on the subject, (but please feel free to post a rebuttal if you wish), but I disagree completely with your statement above.

There is just too much randomisation in the occurrence of events in baseball (in fact, in real life) to exclude the need of some sort of articificial adjusters.

Consider this example: imagine Strat reproduces at 100% all the stats, from the number of times Giants players reach bases to the number of hits allowed/outs performed by opponents. So they exactly reproduce that, say, 300 Giants players were on bases when Sandoval came to hit, that Sandoval hit for .330, and that opponent outfielders retired 15 runners at home plate. But imagine this freak event: the 15 runners who got out at home plate were running after Sandoval's hits. On average, it should be 2; that's 13 rbis that Sandoval "lost" compared to what is expected, despite all events being accurately mesured. There is not just no way to control for this, unless to do play-by-play repetition (which Strat does not do).

These freak events happen all the team. One player gets three grand slams out of his 10 homeruns; another hitter goes .100 with bases loaded, but hits well otherwise in clutch. His clutch numbers will look good, but not his rbi production. etc. etc.

[voovits]

Just to note, this is directly from the CDROM help file

"BALLPARK EFFECTS AND CLUTCH HITTING (Located in the Options/Rules dialog)

These rules add more realism to the game. If you wish to use them (we highly recommend that you do), then we suggest that you check all boxes on.

NOTE: If you wish to use Weather Effects the Ballpark Effects must be checked on.

Ballpark rules add ballparks' effects so that, just as in real life, it is easier to hit a home run in Colorado than it is in Houston. Home runs and batting average are both taken into account, and Strat-O-Matic also factors in every player's home ballpark so that accuracy and realism are maintained.

Clutch hitting measures each player's ability to deliver the clutch hit. Strat-O-Matic's exclusive clutch hitting formula accurately measures every player's ability in these key situations. Traditional clutch hitting stats do not comprise enough situations to make them statistically significant. For instance a batter with 500 at-bats might only hit 110 times with runners in scoring position and perhaps 80 times in "close and late" situations. This total number of at-bats is not enough to truly gauge his clutch hitting ability, because in 190 at-bats a .250 hitter can be expected to swing anywhere between .187 and .313 just as a result of random statistical distribution. Strat-O-Matic's exclusive clutch hitting formula measures the outcome of every baserunner that the batter affected during the year. This detailed approach provides a true measurement of the batter's clutch hitting ability. Our system is implemented with 2 outs and runners in scoring position because in real-life that is traditionally the most difficult time to "come through" with a base hit. If the batter fails, there are no more chances to drive in those runners. Most batters will see their average go down in the clutch -- both in real-life and with our system.

Some leagues wish to play with just the ballpark option and not the clutch hitting system. [b:2509cd5aa6]Originally these options were developed to work together[/b:2509cd5aa6], and in previous versions of the game you could not toggle them individually. The option to toggle them independently is now available, however when doing this the [b:2509cd5aa6]statistical accuracy of the game will suffer slightly[/b:2509cd5aa6], perhaps a point or two on the batting average at most.

[b:2509cd5aa6]Strat-O-Matic recommends that you use both ballpark and clutch hitting features together for maximum realism and statistical accuracy.[/b:2509cd5aa6]"

--

So I stand by what I say about how clutch hitting is as much a part of the game as the ballpark effects.

Also, I don't see SOM say anywhere there that clutch hitting is an RBI adjuster. I don't believe that to be true and I'd like to see where it says that in the help file or instruction booklet, or where an employee of SOM has said that.
Clutch hitting is an integral part of the super advanced game.
I have an idea, why don't we just scrap all the rules and play the basic game?
I'm sure there are other site that don't use clutch hitting.

[qksilver69]

LOL Voovits, you're making my point. SOM had so much demand to cut out the Clutch factor that they did just that - their serious players demanded it because they know it's a sham. Clutch is a sophisticated fudge factor for RBIs.

Again, this has been proven over & over again via advanced stats - there is NO real life "clutch factor". Over time, given enough PAs, a hitter's performance in 2-out, man-in-scoring-position scenarios generally mirrors their non-clutch performance.

When confronted with this evidence (sorry I don't have a link, it was years ago this was disclosed), SOM basically confessed in a forum *on their site* that the main purpose of this stat was so that RBI would not be out of whack with actual performance. All other stats (BA, OB, SLG, splits) could be simmed just fine, but RBI fluctuated wildly because of what MW said above - so many variables (batting order, etc.) come into play that RBI, even more than runs scored, is the most difficult output to control for in the SOM environment.

MW, I think you're missing my point - we here don't NEED to perfectly replicate RBI performance. SOM does because they are claiming to be able to sim all pieces of the MLB season accurately. Their credibility demands some fudge factor for RBI.

Our games in TSN (or any other SOM league) don't have the same requirement. We need to be sure that the parts of the performance *under hitter control* are replicated correctly, not that Adrian Gonzalez' 99 RBI are replicated when he has 3 .400 OBP guys in front of him in a TSN league. To get that 99 RBI, in a TSN environment we are sacrificing 10 or 12 singles that Gonzalez would have had in real life, and so nearly every time in a 12-team TSN league, Gonzalez or similar high OPS/low RBI guys will underperform their cards, usually severely unless they are in a hitter environment. What this gives us, and I see this trend all the time with these players, are HR & 2B totals that match real life, but BA totals that are exceptionally low.

So Voovits in this environment just the opposite is true - clutch hitting impairs real-life performance by some of the best hitters in the game by penalizing them for the guys playing around them.

[coyote303]

I don't think you're giving SOM credit. A player doesn't end up with negative clutch because they don't have opportunities to drive in runs due to a poor supporting cast--they end up with negative clutch because they do poorly with the RBI opportunities they had, however many that may have been.

I'm not saying the clutch system is perfect, but it adds another interesting layer to the game when drafting a team.

If a player got more RBIs than expected because he came through with runners on base, I think it's a cool thing that his SOM card will reflect that.

[Paul_Long71]

I would love to see the percentages of rolls off hitters/pitchers card by each player to see if someone is underperforming or just getting unlucky rolls off the other guy's card.

[Jimbo123]

You're all super experienced veterans and I only have a few seasons under my belt, but here's what I think.

Overall, I like the clutch hitting rating system. It influences who I draft and how I set up my lineups. I enjoy having to weigh the pros and cons of batting a particular hitter in a certain spot in the order. I think it's part of the game's strategy and I hope it stays.

I would really like to see lineup cards for individual games. In almost every league that I play there are one or more managers that set up their starters so you're facing a split featuring RHP/R and RHP/L. It's a clever tactic but I don't care for it. Why should a pitcher get a free pass when facing a good team merely because he was fortuitous enough to have the opposing manager structure his lineup to counter the other starter? Individual lineup cards are a change that's desperately needed.

I don't think the closer option should be changed. Most teams do rely on a single closer.

As far as the % of dice rolls on the hitter's card vs. the pitcher's card, I'm indifferent. Even if your superstar hitter is getting a majority of rolls on the hitter's card and he's still drastically under performing, eventually it will even out, providing that you're taking ball park effects and opposing pitching into account. The bottom line is that, after taking the aforementioned variables into consideration, a good card is a good card. Since players could figure this % split out by doing a game by game analysis, I don't see any harm in including it, but I don't think it's all that valuable of a tool.

Well, that's my two cents.

[coyote303]

[quote:21bd156980="qksilver"]

[b:21bd156980]1) Clutch optional.[/b:21bd156980]...

[b:21bd156980]2) Individual pitching matchup capability.[/b:21bd156980]...

[b:21bd156980]3) Closer optional.[/b:21bd156980]...

[b:21bd156980]4) Pitcher/hitter split roll summary.[/b:21bd156980]...

[/quote:21bd156980]

My two cents...

1. I already posted my opinion on this, so I'll sum up and say I like clutch. Also, too many options can make it more difficult to fill leagues.

2. Giving the manager more control certainly makes sense. However, I prefer the way it is now. Why? Because it helps minimize unrealistic platooning. Someone stacking a bunch of left-handed hitters against a left-handed reverse starter just doesn't sit well with me.

3. I think the whole real-life concept of closers is silly. However, I do think that SOM's closer rules helps ensure realistic use of closers. My only concern is, again, having too many league options. Just having the DH or no-DH option (which is a good option to have IMHO) essentially cuts your new league's potential player pool in half.

4. This idea is harmless, even interesting. However, there are other things I would prefer be added first, such as games played for hitters, GIDP, HBP, pitcher's hitting (in non-DH league).

Of course, if option 4 is implemented, I can then see someone asking for how many of each column have been rolled!

[gbrookes]

For what it's worth, I enjoy the clutch hitting aspect of the game. I like the extra strategy element.

When strat designs the cards, I believe they take into account the team they were playing on in order to achieve probably accuracy with the actual game results. So if the clutch factor IS an RBI adjuster, what's wrong with that, if it accurately mimics real life performance - taking into consideration the strength or the weakness of the team they were playing on, since that is how the strat simulating works, I have to believe. (If it didn't work this way, their ability to sim real life performance for real life teams would NOT work at all, for any hitter characteristics).

Moreover, the players that seem to be consistently be rated as clutch players TEND to have one thing in common - they are contact hitters who tend to hit a lot of groundballs, and strike out rarely. My interpretation of the substantive reality (accuracy) of this trend is that those players ARE ABLE TO MOVE RUNNERS when they need to, by making contact and hitting the ball to the right place to move the runners. What's wrong with rewarding them for that ability? So, if I am right on this point, it means the game mechanism isn't exactly right, since it occurs with 2 outs, but my theory would mean that it occurs with no outs or 1 out in real life, but even so, I still like the rule as a proxy for real life RBI ability.

So, for example, DeJesus has played for KC for a fairly long time. I don't think that is a great lineup (others may disagree). Nevertheless he is consistently rated as a great clutch hitter producing "more than the otherwise expected" amount of RBIs. He IS a good contact hitter, who puts the ball in play, and gets runners to score more than expected from his hitter characteristics would otherwise indicate. Why NOT let the game reflect that?

If there is really something not working according to the goal of real life simulation, why not fix it instead of throwing it out altogether? If we can put a man on the moon....

Cheers!

Geoff Brookes

[gbrookes]

Here is a brain teaser question -

IF the clutch hitting is a "sham" - i.e. given sufficient actual at bats, any player's RBI production would tend to be predicted and replicated PURELY based on their hits and normal player characteristics - then why does a player like DeJesus exist (there are others like him) who are CONSISTENTLY rated as clutch hitters by strat?

In effect, doesn't this prove that some clutch element really DOES exist in the data, that is not PURELY reflected by hits and normal player card elements???

Geoff Brookes

[Jake Squid]

Clutch hitting certainly exists in the past, however past performance is not predictive in this case.

We can look through 2009's data and see who hit well in the clutch and who did not. This is meaningless going forward, though.

Strat is all about replicating past performance, not predicting future performance. That being the case, I have no problem with their clutch ratings.