Deciding lineups based on sums Topic

agentverde
Posts: 164 (0)
Hall of Famer
198 2 70
Prospect
I'm trying to figure out my lineup for the playoffs, and I have a guy who's mediocre against lefties, but still a good contact/power/eye guy, and another who's really good against lefties, but not as strong in the other categories.  I thought, "What if I just add up a player's Contact, Power, Appropriate Split (depending on the pitcher) and Batting Eye?"  Whoever has the highest sum, that's who I go with (of course, speed and fielding factor in, but I'm talking solely about hitting right now).

Maybe that sounds obvious to everyone, but does anyone else do this?  Is it effective?  How do others determine which guy to put in?
8/17/2010 11:53 AM
dedelman
Posts: 2238 (0)
Hall of Famer
324 219
Pro
I have a formuIa; like yours, it is different depending on the opposing SP's handedness.  I have used a number of hitting formulae over the years, but my current one is closest to simple addition of any that I've used.  I think your approach would work OK.
8/17/2010 12:08 PM
mhulshult
Posts: 982 (0)
Hall of Famer
5 3 63
Prospect
If you post the players we can give you some specific analysis.

I would hesitate using this strategy as it causes certain information to get lost in the averages.  If contact, power, split, eye is 45, 95, 95, 50, you have a very different player than 95, 45, 50, 95, or even 65, 75, 70, 75.  They all add up to 285, but you'd want to use them in different ways or combined with different players.  And, as you mentioned, you can't ignore speed, base running, or defense.
 
8/17/2010 12:46 PM (edited)
MikeT23
Posts: 58012 (0)
Hall of Famer
87 1 47 1 189
Prospect
I wouldn't discount speed and baserunning so quickly from your formula.  They are parts of hitting.

Beyond that, you really need to account for ballpark specs.
8/17/2010 1:11 PM
moosep
Posts: 1496 (0)
Hall of Famer
12 1 1 24 2 36
Pro
I go a lot on OBP. I believe the proof is in the pudding.
8/17/2010 7:56 PM
iain
Posts: 1864 (0)
Hall of Famer
35 1 1 18 1 279
Prospect
I'd think it has a lot to do with where in your lineup he'd be going.  Sums might be useful on a macro scale, but the playoffs aren't decided on a macro scale.  If you need a guy to get on and steal a bag, or play solid D, or hit in the middle of the order, you need to use the guy that fits that description.

Far too much noise in a simple addition like that.
8/17/2010 11:04 PM
This post has a rating of , which is below the default threshold.
Show
MikeT23
Posts: 58012 (0)
Hall of Famer
87 1 47 1 189
Prospect
I have a formula for hitters that runs almost perfect for OPS vs L/R.   It's not opinion-based.  It's result-based. 

I won't give it out but a player has to have at least 100 AB vs. the handedness before I'll count it.    Out of 10 players who'll qualify, there are usually one overperformer and one underperformer.   The rest run 1-10 in order. 
8/18/2010 8:50 AM
goyankees2
Posts: 17 (0)
Rookie
3 4
Prospect
I just ran a multiple linear regression -- predicting players' OPS by their batting attributes. I only had a half-season of data, but I still got an R^2 of 58%, which is pretty solid considering it ignores park factor, platooning, etc. It also isn't as advanced statistically as it could be, although I will say it's not just using each attribute as a variable straight-up. I made a little calculator and plan on using it regularly, especially for the draft.
8/24/2010 3:03 PM
mhulshult
Posts: 982 (0)
Hall of Famer
5 3 63
Prospect
I did this for pitchers.  Let me know how it works out for you, but based on the variables I put in, I found the results too inconsistent.  For instance, with starting pitchers, I estimated their opposing OBP, and the ratings I used included their third pitch.  The t-value on the third pitch was 2.5 (telling me it has a significant effect on the result), but the coefficient was positive.  So as the rating of their third pitch goes up, so does their opposing OBP.  It didn't make sense.  And when I put it into practice, it was just too far off of what actually happened.  I'm still tinkering, but haven't been too successful yet.

I suppose it's possible that there are too many other variables (ballpark, defense, pc of catcher) to approximate this for pitchers.  I'll have to give it a go with hitters.

As a side note, it's nice to know I'm not the only one who does this.  My wife made fun of me last night as I told her I was taking ratings of fake players who have accumulated fake stats to see if I can approximate what fake stats they're going to put up next year!
8/24/2010 3:21 PM
goyankees2
Posts: 17 (0)
Rookie
3 4
Prospect
I was gonna do pitchers pretty soon (I literally just did hitters a day ago), but I would be less confident. Any basic pitcher stat on Player Search (how I'm getting the data) is just too dependent on other factors, as you said. 

A potential solution to your problem: either a) ignore 3rd-5th altogether, since they probably aren't that important, or b) use dummy variables to indicate whether a pitcher HAS a third pitch. Correlation does not equal causation -- my guess is that you're getting your result because usually only starters have a 3rd pitch, and that starters are usually worse per inning than relievers. Assign 0 if a pitcher has no 3rd pitch, and 1 if he does. Might fix your problem.

I'm not new to stats, (or WhatIf, as I played SimLeague a while back on an account I no longer have the email for), but I am new to HBD, so this stuff is interesting to me as well.

8/24/2010 3:51 PM
tommy_cian
Posts: 289 (0)
Hall of Famer
3 3 42 68 137
Prospect
I think for creating valuable models to predict pitching attributes you would need to seperate those with more than XX pitches per game vs those with less, say 4 innings.  This would give you two models, one for starters and one for relievers.

If you attempt to predict WHIP, it would be a great indication of performance, but I think you would need to factor in what their home park was, which wouldn't be to difficult (you would need to just make a table of each park, and give each an estimate of whether its a hitters, pitchers or neutral park, then enter two dummy (1=yes, 0=no) variables into the model of hitters and pitchers (do not add three, that skews the model).

Just an FYI, I am the director of analytics for a major consulting firm, I have a team of 27 stats that report to me.  I was education and trained as a statistician, but don't really do the hands on work anymore, but I may break out SAS today and see if I can toy with this...
8/24/2010 4:01 PM
tommy_cian
Posts: 289 (0)
Hall of Famer
3 3 42 68 137
Prospect
Posted by goyankees2 on 8/24/2010 3:51:00 PM (view original):
I was gonna do pitchers pretty soon (I literally just did hitters a day ago), but I would be less confident. Any basic pitcher stat on Player Search (how I'm getting the data) is just too dependent on other factors, as you said. 

A potential solution to your problem: either a) ignore 3rd-5th altogether, since they probably aren't that important, or b) use dummy variables to indicate whether a pitcher HAS a third pitch. Correlation does not equal causation -- my guess is that you're getting your result because usually only starters have a 3rd pitch, and that starters are usually worse per inning than relievers. Assign 0 if a pitcher has no 3rd pitch, and 1 if he does. Might fix your problem.

I'm not new to stats, (or WhatIf, as I played SimLeague a while back on an account I no longer have the email for), but I am new to HBD, so this stuff is interesting to me as well.

or create an ordinal variable, 0 = no 3rd pitch, 1 = third pitch with rating under XXX, 2 = third pitch with rating over xxx.  To figure out what XXX equals, try to graph the stat against the dependant variable and see if there is a clear break in the function (a point where the slope of the line changes, you may even find two breaks). 

8/24/2010 4:03 PM
mitchrapp
Posts: 807 (0)
Hall of Famer
106 20 14 109
Prospect
Posted by MikeT23 on 8/18/2010 8:50:00 AM (view original):
I have a formula for hitters that runs almost perfect for OPS vs L/R.   It's not opinion-based.  It's result-based. 

I won't give it out but a player has to have at least 100 AB vs. the handedness before I'll count it.    Out of 10 players who'll qualify, there are usually one overperformer and one underperformer.   The rest run 1-10 in order. 
Come on, be that guy. Be the tell formula guy.
8/24/2010 4:13 PM
jvford
Posts: 1102 (0)
Hall of Famer
1 15 71
Prospect
Posted by goyankees2 on 8/24/2010 3:03:00 PM (view original):
I just ran a multiple linear regression -- predicting players' OPS by their batting attributes. I only had a half-season of data, but I still got an R^2 of 58%, which is pretty solid considering it ignores park factor, platooning, etc. It also isn't as advanced statistically as it could be, although I will say it's not just using each attribute as a variable straight-up. I made a little calculator and plan on using it regularly, especially for the draft.
The formula I use to predict OPS gave me an R^2 of 79% using career data (while ignoring players with significant ABs before full development or during sharp decline). 

I've had very little luck producing anything of real value for pitching.
8/24/2010 4:34 PM
123 Next ▸
Deciding lineups based on sums Topic
△Top△Bottom

Search Criteria

Terms of Use Customer Support Privacy Statement

© 1999-2026 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.