How sophisticated is defense?

[jte87]

I'm curious about how many factors the AI accounts for on the defensive side while a game is being played? Does it care if Torii Hunter is in CF, compared to someone with less range? With the involvement of DIPS, this sort of defensive measure should be incorporated, since a good defense can help its pitchers BABIP. I think my team has some good defenders, but I can't tell if they are helping me the way I think they should. My 3B is supposed to be a good defender from his scouting report, but it seems like there are alot of GBs to the left side that he "just couldn't reach".

All that is obvious while playing the games is errors of course, and I hope that it is beneficial to fill out a roster with guys that have skills at things besides catching the ball.

My intuition says this game is good enough that it would have to incorporate this, and the scouting reports indicate that it is aware of range, but some confirmation would ease my mind.

[obaslg]

I don't think anyone has studied this in v.6. From the studies I did on v.5, defense was huge - much bigger than I would have thought. I believe that Markus said that he was downgrading defense for v.6. We won't know until we study it, and it's a pretty time-consuming study.

[darrylr]

I would like to test the effects of fielding range/pct/arm but haven't been able to yet. I know I made sure to draft guys with decent fielding attributes in my fictional league and have been successful.

-Darryl

[Biggie Fries]

Defense <i>was</i> huge in OOTP5. I found, in a fictional solo league, that surrounding average pitchers with fielders with great range (who were often otherwise cheap, assuming they weren't abover-average hitters) was enough to keep me ahead of the machine. My staff ERAs were sufficiently below league average, but the pitchers weren't costing me that much and were easy to find.

In fact, it got to the point that when I was looking for players, I would only search for fielders with A or B ratings.

Having been advised that this is less of an advantage in OOTP6, I am trying to "loosen up" somewhat, and live with guys with range factors of 3 or even 2. Funny, but a "3" doesn't look as bad to me as a "C"!

[obaslg]

I would like to test the effects of fielding range/pct/arm but haven't been able to yet. I know I made sure to draft guys with decent fielding attributes in my fictional league and have been successful.

I did a regression analysis of the effect of players' range and arm on total chances, and found no statistically significant impact. I haven't done league-wide studies - it may show up there.

[scotto313]

WHAT?!? Downgrading defense? If we are buying into DIPs, where is the logic in downgrading defense? If there is such a thing as DIPS, then there must be Defense Dependent pitching Stats. The implementation of a DIPS engine should mean an increase in defense importance. I was hoping that the new engine would put more emphasis on defense, including 1B. What is the goal here? If the pitchers have little or no control over balls in play and the defense has less control than before, are we not tipping the scales too far towards the batter.

[Night2000]

Originally posted by scotto313
WHAT?!? Downgrading defense? If we are buying into DIPs, where is the logic in downgrading defense? If there is such a thing as DIPS, then there must be Defense Dependent pitching Stats. The implementation of a DIPS engine should mean an increase in defense importance. I was hoping that the new engine would put more emphasis on defense, including 1B. What is the goal here? If the pitchers have little or no control over balls in play and the defense has less control than before, are we not tipping the scales too far towards the batter.

Good valid points! I'm curious to the answer myself.

[sixto]

I think the logical assumption is that defense was not downgraded in importance so much as integrated into the pitcher-batter confrontation. I suspect that in earlier versions, good defensive players were making plays irrespective of pitcher or hitter ability, and that now, pitcher quality depends on their ability to contain BB, SO and HR, and the primary confrontation after that is between batter and fielder.

Bottom line, however, is that you really can't argue against how the game "values" defense unless you know what its previous value was.

[darrylr]

I don't think that he means that the effect of defense has been taken away but rather the percentage difference between having average fielders and gold glovers isn't as huge as it was in OOTP5. It sounds like it was a little over the top and the main way to win was to have a top defense first. It may have been balanced a little more now.

-Darryl

[Joshv02]

Originally posted by scotto313
WHAT?!? Downgrading defense? If we are buying into DIPs, where is the logic in downgrading defense? If there is such a thing as DIPS, then there must be Defense Dependent pitching Stats. The implementation of a DIPS engine should mean an increase in defense importance. I was hoping that the new engine would put more emphasis on defense, including 1B. What is the goal here? If the pitchers have little or no control over balls in play and the defense has less control than before, are we not tipping the scales too far towards the batter.

I think the issue was more that certain defensive range differences were way too important. So, Jason Moyer did an excellent study that showed A range at SS and 2B were worth ~11 wins vs. E range http://www.ootpdevelopments.com/boar...&threadid=2310 . This seems a bit out of wack.

For comparison, the better Win Share defensive seasons are something like 15ish WS per 162 games (Cordero a few years ago; Neifi Perez last year, extrapolated out, IIRC). That works out to 5 Wins. The best UZR folk are at about 30 runs/year, or 3 wins (the worst of the worst are about negative 30-40 runs, or minus 3-4 wins; the largest possible spread here is 7 wins). Other defensive metrics, like Pinto's Probabilistic method ( www.baseballmusings.com ) show even a tigher spread.

I agree, Scott. 1B defense had to be improved, and hopefully overall defense was improved using some of the newer analyses.

[obaslg]

If we are buying into DIPs, where is the logic in downgrading defense?

To be clear: I have no idea if defense was downgraded - I just thought I read that somewhere. That first study I did didn't show a correlation, but I never ran that particular study in v.5, so I don't know if that's a change.

While it would seem a little strange to me to downgrade defense while changing to DIPS, there could have been a judgment that the old system's defense was way too high. Or the answer could be as suggested, that defense is equally or more important than before, but there's a flatter curve in defensive ability, and chance is more of a factor overall.

[Malleus Dei]

Originally posted by scotto313
If we are buying into DIPs...

You don't "buy into" DIPS any more than you "buy into" quantum mechanics. DIPS is a valid model whether you "buy in" or not.

[sporr]

I think Markus said that defense would have a bigger impact in the correct areas whereas before the impact wasn't in the correct areas.

EDIT: Here's the quote:

"Also, thanks to research in the area, defense now actually plays the role it does in real life In previous version, it was overrated and influenced the results too much and in a wrong way... this has been drastically changed and now works the way it should."

[Crapshoot]

Originally posted by Malleus Dei
You don't "buy into" DIPS any more than you "buy into" quantum mechanics. DIPS is a valid model whether you "buy in" or not.

What he said- sadly, now and then- Mal's correct.


More so, the problem is that in OOTP4/5 defense was ridiculously overrated to the point where a difference between A and E rated SS was forth 8-10 wins- Jason Moyer's study on the subject is probably the best I've seen. DIPS value's defense more accurately.

[obaslg]

More so, the problem is that in OOTP4/5 defense was ridiculously overrated to the point where a difference between A and E rated SS was forth 8-10 wins- Jason Moyer's study on the subject is probably the best I've seen.

I did a study with more specificity, but didn't publish it b/c I'm a competitive wanker, and I didn't want anyone from my league to know. The difference between an A and an E at SS was 60 runs per year. At CF, it was 48 runs per year.

At catcher, it was 36 runs per year - that's range, not arm. In other words, an A range catcher was about 72 RBI better than an E range. Arm meant much less for catcher. I never did understand why catcher's range mattered so much.

[Crapshoot]

My mistake- that correlates to about 6 wins in 5, which is still pretty extreme///

[rdklein]

I'm trying to figure out what significance holding a runner on makes.... one inning I'll hold the runner and he'll steal a base... couple innings later I'll forget to hold him on, and my catcher will throw him out. This has happened a few times already.

Fast runners getting thrown out when they are not held on first.... hmmm?

[scotto313]

I still think the world is flat.