The Effect of Speed Rating on Hits Topic

Very interesting topic to me but I don't have much research on it...

How much of an impact does Speed Rating have on hits? We know (EDIT: at least, I thought we knew) it impact things like infield singles and going first the third on a single, scoring from second, tagging up, etc. Is that effect quantifiable, though and, if so, has anyone compiled data on it?

EDIT: I guess I should have included the point about double plays. Speed rating impacts GIDP, significantly, as well. This is known.
5/9/2016 6:12 PM (edited)
It has no effect on hits, per se. Batting average is what it is, and speed doesn't move the needle on that. It also doesn't seem to have a mechanism, in this game, for altering the probability of an error or a +/- play, although in real life it's hard to argue that more infield errors, at least, occur on plays where the baserunner has speed. Also, the probability of going first to third, tagging up, etc. doesn't seem to be impacted substantially by speed. That's not to say it doesn't matter at all, but it's not huge.

The impact on double plays does seem to be very substantial, however. Off the top of my head, I would say that on average a guy with 80+ speed, all else being equal, hits into fewer than half the double plays of your average catcher.
5/9/2016 5:19 PM
The logic tree:

All of the information in this post comes from a presentation that Paul Bessire at WIS delivered in 2009. You may still be able to find this online somewhere. If anything has changed since 2009, what is listed here may no longer be accurate.

Step 1: Determine if an unusual event occurs (defined as IBB, WP, PB, SB, CS, SH, H&R, Balk, Pickoff) or if it is a normal PA
Step 2: If normal PA, determine if HBP or not
Step 3: If not HBP, determine if walk or not. At this point, if the PA does not result in a walk, then we have an "at bat"
Step 4: Determine if the at bat results in a hit or an out (included park adjustments and any platoon advantage)

If the result of an at bat is an out, then do the following:
Step 5: Strikeout, or "normal" out (and if it's a normal out, is it a groundout or flyout, and in which direction)
Step 6: If "normal" out, determine if an error occurs (fielding rating) or if it becomes a hit (range rating, minus plays)

If the result of an at bat is a hit, then do the following:
Step 7: Determine if HR or "normal - in play" (Park effects used to determine HR)
Step 8: If "normal - in play" then determine if fielder converts it into an out (range, plus play)
Step 9: If it remains a hit, determine what kind of hit: triple, double, single (Park effects used to determine type of hit)

The exact calculations along the way - such as whether an at bat results in an out or a hit - are where WIS's method of Log5 normalization are used. These are also impacted by: platoon advantage, park effects, and how much each outcome is dependent on the batter or the pitcher.

In the same presentation, WIS used the following breakdowns to determine how much the batter and pitcher were responsible for each outcome:

HBP/PA BB/(PA-HBP) H/AB K/(OUT) HR/HIT 2B/HIT 3B/HIT
Pitcher% 47.8 43.5 46.7 45.6 39.7 15.2 11.6
Hitter% 52.2 56.5 53.3 54.4 60.3 84.8 88.4

5/9/2016 5:19 PM
Posted by dahsdebater on 5/9/2016 5:19:00 PM (view original):
It has no effect on hits, per se. Batting average is what it is, and speed doesn't move the needle on that. It also doesn't seem to have a mechanism, in this game, for altering the probability of an error or a +/- play, although in real life it's hard to argue that more infield errors, at least, occur on plays where the baserunner has speed. Also, the probability of going first to third, tagging up, etc. doesn't seem to be impacted substantially by speed. That's not to say it doesn't matter at all, but it's not huge.

The impact on double plays does seem to be very substantial, however. Off the top of my head, I would say that on average a guy with 80+ speed, all else being equal, hits into fewer than half the double plays of your average catcher.
I find it hard to believe that it has no impact on infield hits and things like going first to third, and though I haven't collected data on it, it seems counter to what I've observed as well. Has this been explicitly stated anywhere?
5/9/2016 5:30 PM
Posted by dahsdebater on 5/9/2016 5:19:00 PM (view original):
It has no effect on hits, per se. Batting average is what it is, and speed doesn't move the needle on that. It also doesn't seem to have a mechanism, in this game, for altering the probability of an error or a +/- play, although in real life it's hard to argue that more infield errors, at least, occur on plays where the baserunner has speed. Also, the probability of going first to third, tagging up, etc. doesn't seem to be impacted substantially by speed. That's not to say it doesn't matter at all, but it's not huge.

The impact on double plays does seem to be very substantial, however. Off the top of my head, I would say that on average a guy with 80+ speed, all else being equal, hits into fewer than half the double plays of your average catcher.
And it doesn't know how hard a guy hits the ball. You see a guy like Matt Holliday hit the stuffing out of the baseball all season, yet when he's hitting it right at somebody all the time like he did in '13, he leads the league in GIDP with 31. FWIW, he has a 70 speed rating that season...curious how many he would have in a Sim season...
5/9/2016 5:39 PM (edited)
DoctorKz is right.

Think of it this way - you either set the simulation so that a .300 hitter will hit within a reasonable expected range (a standard deviation statistically) of .300, or you can add speed one way or another and not set it so that he typically hits near .300 more often than not. You can't do both.

You set a .300 hitter to hit .300 and also make speed count and that player is fast, now he is hitting .370 typically, which is not what his speed got him to hit in real life. In real life hit speed, hitting ability and whatever other talents he had made him hit .300. If you set him to hit .260 but add speed hoping it will make him hit .300 with all the infield hits he beats out at first, you run the risk of him hitting .220 too often or whatever.

So the player hit what he hit, including his speed. It is similar to the question asked in another thread about players who hit loooong home runs. Since not every home run is tape measured and even players capable of hitting long ones did not do so in every at bat, and so forth, the only thing that counts is: that player hit 40 homers that year. You either make him typically a 40 homer player, within a certain range of random variation or you make him a 40 homer hitter and give him extra credit for the homers he would have hit (and how could we possibly know how many that would be) in cavernous stadiums. That assumes the player did not already play in such parks and hit his homers there, so we know how many he would have hit already.

It suggests that Mickey Mantle would have MORE homers playing half his games in Dodger Stadium that in the Bronx, since it suggests adding home runs he never hit but which would have been homers in a park he never played in.

WIS does not work that way - it relies strictly on real life stats, normalized to many, but not all factors. Nothing covers all factors anyway (wind speed, what about the home runs that would have been robbed one windy days etc.?).

In Out of the Park Baseball, one has the option playing solitaire (they have online leagues too but I have no idea how they work as I don't play them) to play either some combination of ratings for the players' abilities (speed, gap power, HR power, contact, etc.), and real life stats (and whether you want the latter to be for one season, a variation on the probabilities of three seasons or five, or their whole career), or just one or the other.

So in that game you can play using only ratings, though how those ratings are arrived at I have no idea, but you could go onto their forum (OOTP Development Forums) and ask the makers of the game who are often very active on the site. But here it is real life stats only.
5/9/2016 5:49 PM
I do see a correlation to a guy's speed rating and GIDP. 2000 David Segui always seems to have high GIDP numbers, and his speed rating is...50.
5/9/2016 5:53 PM
I have most certainly seen a strong correlation of speed keeping a hitter from grounding into a twin killing (remember when DPs used to be called "twin killings")?
5/9/2016 6:55 PM
Holliday's 31 GIDP was only a few short of Jim Rice's record of 36 set in 1984. The following season Rice nearly equaled that with hitting into 35. Speed ratings were 67 and 65, respectively.
5/9/2016 7:12 PM
From my take, it's possible that in the Bessire presentation, batter speed affecting infield hits could still be a part of either reversing the "out" determination or of securing the "hit" determination, if there were a result of "infield hit - potential." I mean, it could possibly be that Bessire forgot to mention such a thing.

There has to be some reason that Tim Raines does better compared to his AVG#, than Rogers Hornsby, compared to his, especially when facing the toughest pitchers (which are usually deadball pitchers).

(When I didn't know how pervasive this difference is, whatever makes the difference, it either benefited me unawares, or messed me up, especially in hi-cap leagues.)
5/17/2016 1:36 AM (edited)
Switch-hitters like Raines never have a platoon disadvantage when batting. RHB or LHB facing the same handed pitcher have a 4.5% platoon disadvantage. The 4.5% came from the presentation PP slides, in the Pedro/Ruth example. Given that there are more RHP than LHP, it explains why RHB often appear to come up shorter on sim AVG versus RL avg when compared to switch-hitters or even LHB.

It would be far too long ago to find the post anywhere, but I think Admin or TZ or somebody once clarified that Speed Score is not used to determine SB outcomes, but is used to determine if a batter is successful trying to take an extra base, successful at beating out a DP, and also in determining if a hit is scored as an IF/bunt hit or not. At one point many years ago, players with high speed scores weren't being credit with a fair share of IF and/or bunt hits. A minor engine fix was made to correct that. October 2008 it was. In June 2008 an engine change made speed score a factor in determining the result of sac bunts.
5/17/2016 11:30 AM
I've never heard it said anywhere that speed score could turn a decided out into a hit, like a minus play would turn a defensive out into a hit. I've also never heard it mentioned anywhere that speed score is part of determining hit or out in a normal PA outcome. I would not believe either one to be true.
5/17/2016 11:33 AM
Posted by skunk206 on 5/17/2016 11:30:00 AM (view original):
Switch-hitters like Raines never have a platoon disadvantage when batting. RHB or LHB facing the same handed pitcher have a 4.5% platoon disadvantage. The 4.5% came from the presentation PP slides, in the Pedro/Ruth example. Given that there are more RHP than LHP, it explains why RHB often appear to come up shorter on sim AVG versus RL avg when compared to switch-hitters or even LHB.

It would be far too long ago to find the post anywhere, but I think Admin or TZ or somebody once clarified that Speed Score is not used to determine SB outcomes, but is used to determine if a batter is successful trying to take an extra base, successful at beating out a DP, and also in determining if a hit is scored as an IF/bunt hit or not. At one point many years ago, players with high speed scores weren't being credit with a fair share of IF and/or bunt hits. A minor engine fix was made to correct that. October 2008 it was. In June 2008 an engine change made speed score a factor in determining the result of sac bunts.
Tails up, for skunk!
5/17/2016 10:44 PM
Voice of reason resonates, don't it?
5/17/2016 11:25 PM
The Effect of Speed Rating on Hits Topic

Search Criteria

Terms of Use Customer Support Privacy Statement

© 1999-2024 WhatIfSports.com, Inc. All rights reserved. WhatIfSports is a trademark of WhatIfSports.com, Inc. SimLeague, SimMatchup and iSimNow are trademarks or registered trademarks of Electronic Arts, Inc. Used under license. The names of actual companies and products mentioned herein may be the trademarks of their respective owners.