Study #2 - Defense vs. Hitting

Jason Moyer · 05-22-2002, 02:40 PM

Here's my second study using all-average ratings. Nothing in this study changed except for the defensive ratings at one position on one team.

Difference between A/A and E/E at each position:

SS 10.9 Wins
2B 10.7 Wins
3B 4.8 Wins
RF 4.7 Wins
CF 4.1 Wins
LF 3.1 Wins
C 1.7 Wins
1B -0.8 Wins (yes, the team with an E rated 1B actually faired better against average competition than the team with an A rated first baseman - I'd say 1B defense is unimportant)

Here is the difference between a player who can hit twice as well as the league average vs. one who hits half as well as the league average, all other factors the same:

10.5 Wins

So, in essence, a light hitting A-rated 2B or SS with a small salary will contribute as much to your team's success as a heavy hitting E-rated 2B or SS with a large salary.

What do I take from this? Stick with A rated middle infielders and spend your money on offense elsewhere.

My ultimate compromise lineup, assuming you have to choose between offense/defense at each position and can only afford 3 good hitters (the amount each players wants for a 7 year contract is in parenthesis):

1B - good offense, E defense (4.4mil)
C - good offense, E defense (7.0mil)
LF - good offense, E defense (3.5mil)
CF - average offense, C defense (0.9mil)
RF - average offense, C defense (2.4mil)
3B - average offense, C defense (1.0mil)
2B - poor offense, A defense (0.4mil)
SS - poor offense, A defense (0.4mil)

Cost - $20mil
Wins/Season - 95.8

Not bad when you consider you have average pitching.

Jason

[ 05-22-2002, 08:45 PM: Message edited by: Jason Moyer ]

JAttractive · 05-22-2002, 03:17 PM

I am surprised RF is the highest of the outfield positions though perhaps that would average out more in longer tests. I figured CF would have been third most important but I am not surprised defense means so much for 2b and SS.

Falcon52 · 05-22-2002, 06:36 PM

The difference between A rated fielders and E rated fielders seems significant. My question though is this: What is the difference between having A rated middle infielders and B rated middle infielders?

The reason for this is my SS and 2B are both rated B defensively, but both hit the ball a tonne (my SS has won the MVP award 7 times).

Jason Moyer · 05-22-2002, 07:50 PM

Assuming it would be a linear scale, B defense would be worth about 2.5 wins (7.5 over E defense) and if the guy is twice as good as the average hitter there's another 5 wins (or 10 over a below average hitter).

Jason

Whistler · 05-22-2002, 10:12 PM

Jason, these studies are fascinating. And, like you say, only support the idea that the game is extremely realistic.

Other potential studies: Team leadership. Does it matter?

Create young prospects with exact same ratings and talent ceilings. What percentage turn out to be superstars? Busts?

For pitchers, are some "8" ERA pitchers better than other "8" ERA pitchers? In other words, what sub-groups are most important? Average? Walks? (those would be my best guesses).

Anyway, just some more thoughts. Thanks for the interesting reading.

Killebrew · 05-22-2002, 11:48 PM

Nice jorb Jason & thanks for sharing the results. How many seasons did you sim to get these win total comparisions? Was it just the same backed up season resimmed over and over with 1 defensive range changed each sim? Any details of how you set up the test would be much appreciated

.

Aussie Mark · 05-23-2002, 02:20 AM

<blockquote>quote:<hr />Originally posted by Whistler:
For pitchers, are some "8" ERA pitchers better than other "8" ERA pitchers? In other words, what sub-groups are most important? Average? Walks? (those would be my best guesses).<hr /></blockquote>I haven't tested anywhere near as thoroughly as Jason, however my experience since first playing this game as OOTP2 suggests that avoiding HR and avoiding Walks are the two ratings to look for in pitchers. I don't think this is a secret. Control and the ability to keep the ball in the park are big factors on how many runs a pitcher will give up.

Kemp · 05-23-2002, 02:36 AM

Questions:
Were the results based on 162 game schedule?
Did the player play that position for every inning of every game?

I was thinking about trying to set up something that would allow me to calculate Win Shares for the players in my league. However, you need a ton of data for the defensive WS and I'm starting to thing it isn't worth it.

I'm thinking I could just use Jason's figures to estimate defensive Win Shares based on the number of games a player played at a postion and his range rating

CBL-Commish · 05-23-2002, 02:47 AM

<blockquote>quote:<hr />Originally posted by Jason Moyer:
Here's my second study using all-average ratings. Nothing in this study changed except for the defensive ratings at one position on one team.

Difference between A/A and E/E at each position:

SS 10.9 Wins
2B 10.7 Wins
3B 4.8 Wins
RF 4.7 Wins
CF 4.1 Wins
LF 3.1 Wins
C 1.7 Wins
1B -0.8 Wins (yes, the team with an E rated 1B actually faired better against average competition than the team with an A rated first baseman - I'd say 1B defense is unimportant)

Here is the difference between a player who can hit twice as well as the league average vs. one who hits half as well as the league average, all other factors the same:

10.5 Wins

So, in essence, a light hitting A-rated 2B or SS with a small salary will contribute as much to your team's success as a heavy hitting E-rated 2B or SS with a large salary.

What do I take from this? Stick with A rated middle infielders and spend your money on offense elsewhere.

My ultimate compromise lineup, assuming you have to choose between offense/defense at each position and can only afford 3 good hitters (the amount each players wants for a 7 year contract is in parenthesis):

1B - good offense, E defense (4.4mil)
C - good offense, E defense (7.0mil)
LF - good offense, E defense (3.5mil)
CF - average offense, C defense (0.9mil)
RF - average offense, C defense (2.4mil)
3B - average offense, C defense (1.0mil)
2B - poor offense, A defense (0.4mil)
SS - poor offense, A defense (0.4mil)

Cost - $20mil
Wins/Season - 95.8

Not bad when you consider you have average pitching.

Jason<hr /></blockquote>I'd like to see some more details about length of the study, maybe some other things I can't think of off hand... because if this is true, and I'm looking at this right (which I may not be), then OOTP4 vastly overvalues the effects of defense.

Using Bill James' new Win Shares system the best defensive players of all time contribute far less than this study implies. In Brooks Robinson's best defensive season he was credited with 7.2 defensive win shares, which equates to 2.4 wins. Willie Mays - 8.9 w/s or just under 3 wins. Ozzie Smith - 10.5 w/s or 3.5 wins. The best seasons by the best defensive players of all time are worth no more than 3.5 or 4 wins. Assuming an 'E' rated defender contribues 0, this seems to imply that OOTP overvalues defense by about 3 times.

A shortstop's defense adding 10 wins would be the equivalent of 30 win shares, or by itself a near-MVP caliber season. For example, last year Bret Boone had 32 win shares, Mike Cameron 29, Luis Gonzalez 37 - that's defense, hitting, the whole package. Using these assumptions the 2001 Rich Aurilia would probably be as valuable as the 2001 Barry Bonds.

Plus, this study implies that OOTP has the defensive spectrum wrong. Second basemen aren't more valuable than center fielders. Third basemen aren't more valuable than center fielders.

Think about this logically... Scoring runs is half the game, preventing runs is half the game. Pitchers have to be responsible for a lot of the "preventing runs." So if you assume that a shortstop is worth 1/3 of the team defense (a very high guess, but an easy number to work with) than you'd be saying he's worth about 1/9th of the total team "preventing runs" number. Since preventing runs is half the game, than a shortstops defense couldn't be more than 1/18th of a team's wins. On a great team that equates to something like 5 wins, or not too far off James' estimates.

Am I looking at something wrong here? Am I missing something? Is the study flawed? Am I? I'd really like to hear some comments on this.

Kemp · 05-23-2002, 03:00 AM

You know, you're absolutely right.

When I first looked at those numbers, I compared them to the Win Shares figures and, while they seemed high, it didn't sound completely unreasonable. But I COMPLETELY FORGOT that 1 win = 3 win shares.

So I think your questions are very valid.

[ 05-23-2002, 08:00 AM: Message edited by: Kemp ]

Jason Moyer · 05-23-2002, 07:39 AM

<blockquote>quote:<hr />Originally posted by Killebrew:
Nice jorb Jason & thanks for sharing the results. How many seasons did you sim to get these win total comparisions? Was it just the same backed up season resimmed over and over with 1 defensive range changed each sim? Any details of how you set up the test would be much appreciated

.<hr /></blockquote>I have it set up as a 2-team replay league, no financials, no trades, no injuries, and no fatigue. I reset the season after each year. After the first few tests I was checking ratings to make sure no one developed + or - and apparently in a replay league player development is turned off. (?)

For these tests I just simmed 10 seasons which is a small sample size but I figure 1620 games should give you good tho not perfect figures. Ultimately when I figure out where I want to go with this test league I'll likely pump it up to 100 seasons or more.

Every single player in the league (including pitchers) has hitting ratings set to .257 AVG 24 2B 4 3B 13 HR 53 BB 91 K. All pitchers have the same ratings, except for 3B since there isn't a pitching rating for that. All players are rated only at their primary position with C range and a .976 fielding percentage. In the 10 season control sim, with nothing edited, team A averaged 81.5 wins and team B averaged 80.5 wins. The idea here is to even out everything so that you can measure the impact of a single change or set of changes.

For the hitting study, a strong hitter hit .307 with 48 2B, 8 3B, 26 HR, 106 BB, 46 K, while the weak hitter hit .207 with 12 2B 2 3B 7 HR 27 BB 46 K.

If anyone would like a copy of the test league and can post it somewhere, let me know. It would be useful to have other people doing similar tests since I like doing stuff besides researching probability in baseball games.

Jason

Jason Moyer · 05-23-2002, 07:53 AM

<blockquote>quote:<hr />Originally posted by CBL-Commish:
Using Bill James' new Win Shares system the best defensive players of all time contribute far less than this study implies. In Brooks Robinson's best defensive season he was credited with 7.2 defensive win shares, which equates to 2.4 wins. Willie Mays - 8.9 w/s or just under 3 wins. Ozzie Smith - 10.5 w/s or 3.5 wins. The best seasons by the best defensive players of all time are worth no more than 3.5 or 4 wins. Assuming an 'E' rated defender contribues 0, this seems to imply that OOTP overvalues defense by about 3 times. <hr /></blockquote>That's a good point. Unfortunately I don't have the win shares book yet, but one thing that would be interesting is to run similar tests and at the end of each season go back and calculate the defensive win shares that the player being tested would have received using James method. Maybe OOTP overvalues defense, maybe we're missing so mething obvious when we're looking at the numbers, maybe I ran too few seasons in my tests, or maybe James system isn't as accurate at predicting defensive contribution as he thinks.

Jason

Jim Lynch · 05-23-2002, 08:21 AM

Interesting idea for a study. If these results held up then defense is way out of whack in OOTP.

My question is this, did a player who was worth ten games make a team 91-71 or did they make it 86-76? Either way it would still be crazy.

Just from playing the game, it feels like there is to much of a difference between an A and an E.

It seems that if you took the 30 'starting' centerfielders in the major leagues, you should have 4 a's, 6 b's, 18 c's and 2 d's. What major league center fielder would be a E. Heck I can't even think of who you could say would be a D.

As far as arms are concerned, it would seem 5 a's 4 b's 10 c's, 8 d's and 3 e's.

As far as the importance of position it would seem cf, ss, c, 2b, rf, lf, 3b, 1b... but defense is relative, if you have A. Jones in center you can have statues in right and left. If you have Tino Martinez at first your infielders arms don't need to be as accurate. Baseball might be an 'individual' sport, but in regards to defense, one players 'fielding percentage' can be somewhat related to who is around them.

Jason Moyer · 05-23-2002, 08:28 AM

A team with an A rated 2B or SS won 86 games on average against average competition. A team with an E rated 2B or SS won 76 games against average competition.

Jason

sporr · 05-23-2002, 09:14 AM

I don't think this study suggests that an A shortstop is worth 10 wins. The study says that an A shortstop is worth 10 more wins than the worst shortstops in the league. That "A" is only giving a team 5 more wins than an average SS would. An "E" SS is actually costing his team 5 wins that they would have had if they had an average player there.

Jim Lynch · 05-23-2002, 09:30 AM

Five just seems high to me, at any position.
Not nearly as bad as I first understood (as 10 or 91-71). Look at it this way, if you took the worst outfielder in the major leagues (Greg Vaughn for my money) and replaced him with A perfect LF - keep Vaughn's offensive stats - it might take them from 65 to 68 wins tops. If the worst major league shortstop was made a perfected defensive player, the most they could hope for would be 4 (if everything broke right 5) wins, which would get you from 76 to 80, but not from 76 to 86.

I don't think that many would disagree with that. Instead of average players, how about a replay league where you tweak the ratings the same way. Say the Rolen rosters, take a team (maybe in a park where you feel defense matters more, for infielders on turf, for outfielders huge expanses), play them 10 season as they come. Tweak the player from whatever he is to A/A, then to E/E, see what the difference is in those records. Just an idea

.

Daimyo · 05-23-2002, 09:36 AM

I think it would be interesting to see what sort of diminishing returns effect there is. ie, assuming you have an A range SS, how much does the 2B's defense matter? Will an A range shortstop cover up a poorer 2B (same for an A range CF covering for poor LF/RF)? What if one team was all A range and the other all E range... that would give the maximum defensive advantage (which I assume must be lower than the sum of the indivisual advantages).

sporr · 05-23-2002, 10:01 AM

The other problem with any kind of study with baseball is that there are so many factors involved it's hard to separate them all. There are pitchers with various grounder % for example that will affect the importance of various positions. In addition, you have fielders with the same range who have different fielding %. My guess is that the 10-win total is being bumped up by some other factors. There are probably factors in the game that we aren't even aware of.

Jim Lynch · 05-23-2002, 10:38 AM

Sporr,

That is the nail right on the head. However it seems from what he did do with his study shows that defense might be a little 'overstated' in OOTP. That is just a gut feeling from me, and I could be way off base or I could be right, but I think it's worth trying to figure out. If anything, I'll now look at defense more than I used to when building my teams

.

obaslg · 05-23-2002, 11:11 AM

I disagree with Sporr in a couple of regards.

First, I think there should be E CFs. I think all of the range ratings should be relative to the other guys who play that position, so that there should be a bell curve peaking on C for every position. If they were absolute instead of relative, there shouldn't be any A 1B, b/c no 1B is among the best IFs in baseball. If they are relative, though, it should be reflected in their ability to change position, so that any SS should be able to move to 1B with a good chance of getting a higher rating at 1B, b/c of the relatively weaker fielders there.

As far as the wins go, I think the Vaughn switch sounds a little high (though I've never seen him play), but from watching Rey Ordonez, I have no trouble believing he's worth several wins over the league's worst SS (offense aside, of course).

05-22-2002, 02:40 PM	#1
Jason Moyer Hall Of Famer Join Date: Feb 2002 Location: Pittsburgh Posts: 5,088	Study #2 - Defense vs. Hitting Here's my second study using all-average ratings. Nothing in this study changed except for the defensive ratings at one position on one team. Difference between A/A and E/E at each position: SS 10.9 Wins 2B 10.7 Wins 3B 4.8 Wins RF 4.7 Wins CF 4.1 Wins LF 3.1 Wins C 1.7 Wins 1B -0.8 Wins (yes, the team with an E rated 1B actually faired better against average competition than the team with an A rated first baseman - I'd say 1B defense is unimportant) Here is the difference between a player who can hit twice as well as the league average vs. one who hits half as well as the league average, all other factors the same: 10.5 Wins So, in essence, a light hitting A-rated 2B or SS with a small salary will contribute as much to your team's success as a heavy hitting E-rated 2B or SS with a large salary. What do I take from this? Stick with A rated middle infielders and spend your money on offense elsewhere. My ultimate compromise lineup, assuming you have to choose between offense/defense at each position and can only afford 3 good hitters (the amount each players wants for a 7 year contract is in parenthesis): 1B - good offense, E defense (4.4mil) C - good offense, E defense (7.0mil) LF - good offense, E defense (3.5mil) CF - average offense, C defense (0.9mil) RF - average offense, C defense (2.4mil) 3B - average offense, C defense (1.0mil) 2B - poor offense, A defense (0.4mil) SS - poor offense, A defense (0.4mil) Cost - $20mil Wins/Season - 95.8 Not bad when you consider you have average pitching. Jason <small>[ 05-22-2002, 08:45 PM: Message edited by: Jason Moyer ]</small> __________________ "I pretty much popped everything cold turkey. We were doing steroids they wouldn't give to horses." -- Tom House "I was very fortunate to have a pitching coach by the name of Tom House...Tom, I really miss those days that we spent in the weight room and out on the field working together." -- Nolan Ryan's HoF Induction Speech

05-22-2002, 06:36 PM	#3
Falcon52 All Star Reserve Join Date: Mar 2002 Location: Canada Posts: 615	The difference between A rated fielders and E rated fielders seems significant. My question though is this: What is the difference between having A rated middle infielders and B rated middle infielders? The reason for this is my SS and 2B are both rated B defensively, but both hit the ball a tonne (my SS has won the MVP award 7 times). __________________ "D-FENCE! D-FENCE!"

05-22-2002, 07:50 PM	#4
Jason Moyer Hall Of Famer Join Date: Feb 2002 Location: Pittsburgh Posts: 5,088	Assuming it would be a linear scale, B defense would be worth about 2.5 wins (7.5 over E defense) and if the guy is twice as good as the average hitter there's another 5 wins (or 10 over a below average hitter). Jason __________________ "I pretty much popped everything cold turkey. We were doing steroids they wouldn't give to horses." -- Tom House "I was very fortunate to have a pitching coach by the name of Tom House...Tom, I really miss those days that we spent in the weight room and out on the field working together." -- Nolan Ryan's HoF Induction Speech

05-22-2002, 10:12 PM	#5
Whistler Minors (Rookie Ball) Join Date: Jan 2002 Posts: 30	Jason, these studies are fascinating. And, like you say, only support the idea that the game is extremely realistic. Other potential studies: Team leadership. Does it matter? Create young prospects with exact same ratings and talent ceilings. What percentage turn out to be superstars? Busts? For pitchers, are some "8" ERA pitchers better than other "8" ERA pitchers? In other words, what sub-groups are most important? Average? Walks? (those would be my best guesses). Anyway, just some more thoughts. Thanks for the interesting reading. __________________ Vancouver Whistlers -- Green Monsta League

05-22-2002, 11:48 PM	#6
Killebrew Hall Of Famer Join Date: Dec 2001 Posts: 3,326	Nice jorb Jason & thanks for sharing the results. How many seasons did you sim to get these win total comparisions? Was it just the same backed up season resimmed over and over with 1 defensive range changed each sim? Any details of how you set up the test would be much appreciated. __________________ FOBL Capital City Metropolitans CDL Minnesota Twins

05-22-2002, 03:17 PM	#2
JAttractive All Star Starter Join Date: May 2002 Location: St. Catharines, Ontario Posts: 1,135	I am surprised RF is the highest of the outfield positions though perhaps that would average out more in longer tests. I figured CF would have been third most important but I am not surprised defense means so much for 2b and SS.

05-23-2002, 02:20 AM	#7
Aussie Mark Major Leagues Join Date: Dec 2001 Location: Downunder Posts: 489	</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial">Originally posted by Whistler: <strong>For pitchers, are some "8" ERA pitchers better than other "8" ERA pitchers? In other words, what sub-groups are most important? Average? Walks? (those would be my best guesses).</strong></font><hr /></blockquote><font size="2" face="Verdana, Arial">I haven't tested anywhere near as thoroughly as Jason, however my experience since first playing this game as OOTP2 suggests that avoiding HR and avoiding Walks are the two ratings to look for in pitchers. I don't think this is a secret. Control and the ability to keep the ball in the park are big factors on how many runs a pitcher will give up. __________________ Cheers Mark

05-23-2002, 02:36 AM	#8
Kemp All Star Starter Join Date: Feb 2002 Posts: 1,181	Questions: Were the results based on 162 game schedule? Did the player play that position for every inning of every game? I was thinking about trying to set up something that would allow me to calculate Win Shares for the players in my league. However, you need a ton of data for the defensive WS and I'm starting to thing it isn't worth it. I'm thinking I could just use Jason's figures to estimate defensive Win Shares based on the number of games a player played at a postion and his range rating

05-23-2002, 02:47 AM	#9
CBL-Commish All Star Starter Join Date: Dec 2001 Location: Maryland Posts: 1,999	</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial">Originally posted by Jason Moyer: <strong>Here's my second study using all-average ratings. Nothing in this study changed except for the defensive ratings at one position on one team. Difference between A/A and E/E at each position: SS 10.9 Wins 2B 10.7 Wins 3B 4.8 Wins RF 4.7 Wins CF 4.1 Wins LF 3.1 Wins C 1.7 Wins 1B -0.8 Wins (yes, the team with an E rated 1B actually faired better against average competition than the team with an A rated first baseman - I'd say 1B defense is unimportant) Here is the difference between a player who can hit twice as well as the league average vs. one who hits half as well as the league average, all other factors the same: 10.5 Wins So, in essence, a light hitting A-rated 2B or SS with a small salary will contribute as much to your team's success as a heavy hitting E-rated 2B or SS with a large salary. What do I take from this? Stick with A rated middle infielders and spend your money on offense elsewhere. My ultimate compromise lineup, assuming you have to choose between offense/defense at each position and can only afford 3 good hitters (the amount each players wants for a 7 year contract is in parenthesis): 1B - good offense, E defense (4.4mil) C - good offense, E defense (7.0mil) LF - good offense, E defense (3.5mil) CF - average offense, C defense (0.9mil) RF - average offense, C defense (2.4mil) 3B - average offense, C defense (1.0mil) 2B - poor offense, A defense (0.4mil) SS - poor offense, A defense (0.4mil) Cost - $20mil Wins/Season - 95.8 Not bad when you consider you have average pitching. Jason</strong></font><hr /></blockquote><font size="2" face="Verdana, Arial">I'd like to see some more details about length of the study, maybe some other things I can't think of off hand... because if this is true, and I'm looking at this right (which I may not be), then OOTP4 vastly overvalues the effects of defense. Using Bill James' new Win Shares system the best defensive players of all time contribute far less than this study implies. In Brooks Robinson's best defensive season he was credited with 7.2 defensive win shares, which equates to 2.4 wins. Willie Mays - 8.9 w/s or just under 3 wins. Ozzie Smith - 10.5 w/s or 3.5 wins. The best seasons by the best defensive players of all time are worth no more than 3.5 or 4 wins. Assuming an 'E' rated defender contribues 0, this seems to imply that OOTP overvalues defense by about 3 times. A shortstop's defense adding 10 wins would be the equivalent of 30 win shares, or by itself a near-MVP caliber season. For example, last year Bret Boone had 32 win shares, Mike Cameron 29, Luis Gonzalez 37 - that's defense, hitting, the whole package. Using these assumptions the 2001 Rich Aurilia would probably be as valuable as the 2001 Barry Bonds. Plus, this study implies that OOTP has the defensive spectrum wrong. Second basemen aren't more valuable than center fielders. Third basemen aren't more valuable than center fielders. Think about this logically... Scoring runs is half the game, preventing runs is half the game. Pitchers have to be responsible for a lot of the "preventing runs." So if you assume that a shortstop is worth 1/3 of the team defense (a very high guess, but an easy number to work with) than you'd be saying he's worth about 1/9th of the total team "preventing runs" number. Since preventing runs is half the game, than a shortstops defense couldn't be more than 1/18th of a team's wins. On a great team that equates to something like 5 wins, or not too far off James' estimates. Am I looking at something wrong here? Am I missing something? Is the study flawed? Am I? I'd really like to hear some comments on this. __________________ For the best in O's news: Orioles' Hangout.com

05-23-2002, 03:00 AM	#10
Kemp All Star Starter Join Date: Feb 2002 Posts: 1,181	You know, you're absolutely right. When I first looked at those numbers, I compared them to the Win Shares figures and, while they seemed high, it didn't sound completely unreasonable. But I COMPLETELY FORGOT that 1 win = 3 win shares. So I think your questions are very valid. <small>[ 05-23-2002, 08:00 AM: Message edited by: Kemp ]</small>

05-23-2002, 07:39 AM	#11
Jason Moyer Hall Of Famer Join Date: Feb 2002 Location: Pittsburgh Posts: 5,088	</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial">Originally posted by Killebrew: <strong>Nice jorb Jason & thanks for sharing the results. How many seasons did you sim to get these win total comparisions? Was it just the same backed up season resimmed over and over with 1 defensive range changed each sim? Any details of how you set up the test would be much appreciated.</strong></font><hr /></blockquote><font size="2" face="Verdana, Arial">I have it set up as a 2-team replay league, no financials, no trades, no injuries, and no fatigue. I reset the season after each year. After the first few tests I was checking ratings to make sure no one developed + or - and apparently in a replay league player development is turned off. (?) For these tests I just simmed 10 seasons which is a small sample size but I figure 1620 games should give you good tho not perfect figures. Ultimately when I figure out where I want to go with this test league I'll likely pump it up to 100 seasons or more. Every single player in the league (including pitchers) has hitting ratings set to .257 AVG 24 2B 4 3B 13 HR 53 BB 91 K. All pitchers have the same ratings, except for 3B since there isn't a pitching rating for that. All players are rated only at their primary position with C range and a .976 fielding percentage. In the 10 season control sim, with nothing edited, team A averaged 81.5 wins and team B averaged 80.5 wins. The idea here is to even out everything so that you can measure the impact of a single change or set of changes. For the hitting study, a strong hitter hit .307 with 48 2B, 8 3B, 26 HR, 106 BB, 46 K, while the weak hitter hit .207 with 12 2B 2 3B 7 HR 27 BB 46 K. If anyone would like a copy of the test league and can post it somewhere, let me know. It would be useful to have other people doing similar tests since I like doing stuff besides researching probability in baseball games. Jason __________________ "I pretty much popped everything cold turkey. We were doing steroids they wouldn't give to horses." -- Tom House "I was very fortunate to have a pitching coach by the name of Tom House...Tom, I really miss those days that we spent in the weight room and out on the field working together." -- Nolan Ryan's HoF Induction Speech

05-23-2002, 07:53 AM	#12
Jason Moyer Hall Of Famer Join Date: Feb 2002 Location: Pittsburgh Posts: 5,088	</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial">Originally posted by CBL-Commish: <strong>Using Bill James' new Win Shares system the best defensive players of all time contribute far less than this study implies. In Brooks Robinson's best defensive season he was credited with 7.2 defensive win shares, which equates to 2.4 wins. Willie Mays - 8.9 w/s or just under 3 wins. Ozzie Smith - 10.5 w/s or 3.5 wins. The best seasons by the best defensive players of all time are worth no more than 3.5 or 4 wins. Assuming an 'E' rated defender contribues 0, this seems to imply that OOTP overvalues defense by about 3 times. </strong></font><hr /></blockquote><font size="2" face="Verdana, Arial">That's a good point. Unfortunately I don't have the win shares book yet, but one thing that would be interesting is to run similar tests and at the end of each season go back and calculate the defensive win shares that the player being tested would have received using James method. Maybe OOTP overvalues defense, maybe we're missing so mething obvious when we're looking at the numbers, maybe I ran too few seasons in my tests, or maybe James system isn't as accurate at predicting defensive contribution as he thinks. Jason __________________ "I pretty much popped everything cold turkey. We were doing steroids they wouldn't give to horses." -- Tom House "I was very fortunate to have a pitching coach by the name of Tom House...Tom, I really miss those days that we spent in the weight room and out on the field working together." -- Nolan Ryan's HoF Induction Speech

05-23-2002, 08:21 AM	#13
Jim Lynch Minors (Rookie Ball) Join Date: Mar 2002 Location: Connecticut Posts: 22	Interesting idea for a study. If these results held up then defense is way out of whack in OOTP. My question is this, did a player who was worth ten games make a team 91-71 or did they make it 86-76? Either way it would still be crazy. Just from playing the game, it feels like there is to much of a difference between an A and an E. It seems that if you took the 30 'starting' centerfielders in the major leagues, you should have 4 a's, 6 b's, 18 c's and 2 d's. What major league center fielder would be a E. Heck I can't even think of who you could say would be a D. As far as arms are concerned, it would seem 5 a's 4 b's 10 c's, 8 d's and 3 e's. As far as the importance of position it would seem cf, ss, c, 2b, rf, lf, 3b, 1b... but defense is relative, if you have A. Jones in center you can have statues in right and left. If you have Tino Martinez at first your infielders arms don't need to be as accurate. Baseball might be an 'individual' sport, but in regards to defense, one players 'fielding percentage' can be somewhat related to who is around them.

05-23-2002, 08:28 AM	#14
Jason Moyer Hall Of Famer Join Date: Feb 2002 Location: Pittsburgh Posts: 5,088	A team with an A rated 2B or SS won 86 games on average against average competition. A team with an E rated 2B or SS won 76 games against average competition. Jason __________________ "I pretty much popped everything cold turkey. We were doing steroids they wouldn't give to horses." -- Tom House "I was very fortunate to have a pitching coach by the name of Tom House...Tom, I really miss those days that we spent in the weight room and out on the field working together." -- Nolan Ryan's HoF Induction Speech

05-23-2002, 09:14 AM	#15
sporr Global Moderator Join Date: Dec 2001 Location: Muscatine, IA Posts: 8,277	I don't think this study suggests that an A shortstop is worth 10 wins. The study says that an A shortstop is worth 10 more wins than the worst shortstops in the league. That "A" is only giving a team 5 more wins than an average SS would. An "E" SS is actually costing his team 5 wins that they would have had if they had an average player there.

05-23-2002, 09:30 AM	#16
Jim Lynch Minors (Rookie Ball) Join Date: Mar 2002 Location: Connecticut Posts: 22	Five just seems high to me, at any position. Not nearly as bad as I first understood (as 10 or 91-71). Look at it this way, if you took the worst outfielder in the major leagues (Greg Vaughn for my money) and replaced him with A perfect LF - keep Vaughn's offensive stats - it might take them from 65 to 68 wins tops. If the worst major league shortstop was made a perfected defensive player, the most they could hope for would be 4 (if everything broke right 5) wins, which would get you from 76 to 80, but not from 76 to 86. I don't think that many would disagree with that. Instead of average players, how about a replay league where you tweak the ratings the same way. Say the Rolen rosters, take a team (maybe in a park where you feel defense matters more, for infielders on turf, for outfielders huge expanses), play them 10 season as they come. Tweak the player from whatever he is to A/A, then to E/E, see what the difference is in those records. Just an idea .

05-23-2002, 09:36 AM	#17
Daimyo Bat Boy Join Date: Jan 2002 Location: Evanston, IL Posts: 15	I think it would be interesting to see what sort of diminishing returns effect there is. ie, assuming you have an A range SS, how much does the 2B's defense matter? Will an A range shortstop cover up a poorer 2B (same for an A range CF covering for poor LF/RF)? What if one team was all A range and the other all E range... that would give the maximum defensive advantage (which I assume must be lower than the sum of the indivisual advantages).

05-23-2002, 10:01 AM	#18
sporr Global Moderator Join Date: Dec 2001 Location: Muscatine, IA Posts: 8,277	The other problem with any kind of study with baseball is that there are so many factors involved it's hard to separate them all. There are pitchers with various grounder % for example that will affect the importance of various positions. In addition, you have fielders with the same range who have different fielding %. My guess is that the 10-win total is being bumped up by some other factors. There are probably factors in the game that we aren't even aware of.

05-23-2002, 10:38 AM	#19
Jim Lynch Minors (Rookie Ball) Join Date: Mar 2002 Location: Connecticut Posts: 22	Sporr, That is the nail right on the head. However it seems from what he did do with his study shows that defense might be a little 'overstated' in OOTP. That is just a gut feeling from me, and I could be way off base or I could be right, but I think it's worth trying to figure out. If anything, I'll now look at defense more than I used to when building my teams .

05-23-2002, 11:11 AM	#20
obaslg All Star Reserve Join Date: Dec 2001 Posts: 887	I disagree with Sporr in a couple of regards. First, I think there should be E CFs. I think all of the range ratings should be relative to the other guys who play that position, so that there should be a bell curve peaking on C for every position. If they were absolute instead of relative, there shouldn't be any A 1B, b/c no 1B is among the best IFs in baseball. If they are relative, though, it should be reflected in their ability to change position, so that any SS should be able to move to 1B with a good chance of getting a higher rating at 1B, b/c of the relatively weaker fielders there. As far as the wins go, I think the Vaughn switch sounds a little high (though I've never seen him play), but from watching Rey Ordonez, I have no trouble believing he's worth several wins over the league's worst SS (offense aside, of course). __________________ Realy good musition of many insterments, including the hyperbolic vitriol.