my questions on dips

[Dark Horse]

Originally posted by IatricSB
One hit per game is the difference between a starter and bullpen or AAA. How is that insignificant? I must be missing something.

That is the reasoning on why a pitcher's ability to prevent hits should not be put in the game. The argument is that 1 hit per game is not significant enough.

i was trying to make the argument that 1 hit per game is important.

[IatricSB]

It sounded just the opposite, but I guess you were writing it tongue in cheek and if so, I missed that. Sorry about that.

[clarnzz]

Keep in mind though that one hit per game is between the worst and the best, not between the best and average. It is also based on a single season of data, the same pitcher could be the best on a staff one season and the worst the next.

[Dark Horse]

Originally posted by clarnzz
Keep in mind though that one hit per game is between the worst and the best, not between the best and average. It is also based on a single season of data, the same pitcher could be the best on a staff one season and the worst the next.

Bleacher Bum's study was based on eleven years. And there are other articles in this thread that showed that players consistently performed above or below average in this area throughout their career.

And if you look based on average in this respect the same would go for the other statistics i mentioned. In which case, my premise would be the same. The difference between good and average in this average is not significantly smaller than the difference in other areas like including HR/9, home run ratio, field pct and range factor.

[clarnzz]

Originally posted by Dark Horse
Bleacher Bum's study was based on eleven years. And there are other articles in this thread that showed that players consistently performed above or below average in this area throughout their career.

And if you look based on average in this respect the same would go for the other statistics i mentioned. In which case, my premise would be the same. The difference between good and average in this average is not significantly smaller than the difference in other areas like including HR/9, home run ratio, field pct and range factor.

Oh, I thought you had come up with that one hit per game based on the 1979 Dodgers. I don't insist that such a thing as avoiding hits does not exist, but I do think taking exception type players as the best and worse, does overstate it's significance.

[Skipaway]

Originally posted by Dark Horse
Bleacher Bum's study was based on eleven years. And there are other articles in this thread that showed that players consistently performed above or below average in this area throughout their career.

And if you look based on average in this respect the same would go for the other statistics i mentioned. In which case, my premise would be the same. The difference between good and average in this average is not significantly smaller than the difference in other areas like including HR/9, home run ratio, field pct and range factor.

Hmm, by the same logic, perhaps we should have a RBI rating for hitters too?

We should realize the difference you are seeing in those kind of study are basically a result of ball in play distribution plus the defense skills. As shown by gmo's numbers, the game is already doing that! What we should discuss is how to enhance that.

It's also evident in the Bleacher Bum's thread in Talk Sports. Pitchers like Don Sutton got a sudden shift in BABIP - team BABIP when the shortstop was changed from Wills to Russel. What we should do is to make sure that will happen in OOTP6, and giving Don Sutton an ability to prevent runs won't do that.

It would be as wrong as giving Joe Carter the magical ability to generate more RBIs.

[Dark Horse]

Originally posted by Skipaway
Hmm, by the same logic, perhaps we should have a RBI rating for hitters too?

That opens up a whole different can of worms. i don't believe an RBI rating is necessary. Because i don't believe in clutch hitting.

It's also evident in the Bleacher Bum's thread in Talk Sports. Pitchers like Don Sutton got a sudden shift in BABIP - team BABIP when the shortstop was changed from Wills to Russel. What we should do is to make sure that will happen in OOTP6, and giving Don Sutton an ability to prevent runs won't do that.

If defense in v6 works as advertised there would be a change in BABIP when you replace one defender with a defender of a different rating.

The point is that all of the pitchers on the staff will have a shift in their BABIP. But, with all other things static, the relation in the staff would remain the same. Pitchers with the better BABIP in the first year would still have the better BABIP in the following year with the new defender.

[Dark Horse]

Originally posted by Skipaway
Hmm, by the same logic, perhaps we should have a RBI rating for hitters too?

That opens up a whole different can of worms. i don't believe an RBI rating is necessary. Because i don't believe in clutch hitting.

It's also evident in the Bleacher Bum's thread in Talk Sports. Pitchers like Don Sutton got a sudden shift in BABIP - team BABIP when the shortstop was changed from Wills to Russel.

If defense in v6 works as advertised there would be a change in BABIP when you replace one defender with a defender of a different rating. If one defender had an A rating for range. And the next defender had a C rating. Then A would get to x amount more balls than C would. And the BABIP would change accordingly.

What we should do is to make sure that will happen in OOTP6, and giving Don Sutton an ability to prevent runs won't do that.

i don't want to give a pitcher the ability to prevent runs. i just think a pitcher's ability to prevent hits should be taken into consideration.

The point is that all of the pitchers on the staff will have a similar shift in their BABIP. But, with all other things static, the relation in the staff would remain the same. Pitchers with the better BABIP in the first year would still have the better BABIP in the following year with the new defender.

[BarryZito75]

Originally posted by Dark Horse
That opens up a whole different can of worms. i don't believe an RBI rating is necessary. Because i don't believe in clutch hitting.

I think that skipaway was being sarcastic, and he compared an RBI rating, an insignificant statistic, to an avoiding hits rating, another insignificant statistic.

The funny thing is, as stupid as an RBI rating sounds, hitters probably have more control over RBIs than pitchers do with avoiding hits.

[Skipaway]

Originally posted by Dark Horse

The point is that all of the pitchers on the staff will have a shift in their BABIP. But, with all other things static, the relation in the staff would remain the same. Pitchers with the better BABIP in the first year would still have the better BABIP in the following year with the new defender.

I think this is where you got it wrong. The relation in the staff will NOT remain the same if you got a shortstop change, as shown by the case of Don Sutton.

Sutton was worse than team average with Wills, but better than team average with Russel.

It's that hard an idea to understand, flyball/groundball tendency and left/right-handed of the pitchers would surely change the number of balls in play a shortstop would see. So how can you claim the relation in the staff would remain the same? So what Mitchel Lichtman did by studying pitchers that changed teams actually made much more sense than what Tom Tippett did. Only that might give us a chance to neutralize the defense and park impact on hits allowed.

[Dark Horse]

i'd like to see that study. i've only seen the one Bleacher Bum did in this thread. In that one Don Sutton was always had a better than average BABIP. That's from 1972 to 1980.

[Skipaway]

I think this is the next step for OOTP:

http://www.baseballprimer.com/articl...-02-29_0.shtml

A 2D engine might actually do wonders with these ideas, and give people a better idea of all these DIPS talks.

[Skipaway]

Originally posted by Dark Horse
i'd like to see that study. i've only seen the one Bleacher Bum did in this thread. In that one Don Sutton was always had a better than average BABIP. That's from 1972 to 1980.

Numbers from baseballprospectus:

Sutton's delta H:


1966 0
1967 1
1968 2
1969 5
1970 5
1971 5
1972 -23
1973 -5
1974 -3
1975 -4
1976 -6
1977 -19
1978 -3
1979 -16
1980 -27

Delta-H
The number of hits above or below average for this pitcher, based on his own number of balls in play and his team¡¦s rate of hits (minus homeruns) per ball in play; (H-HR) - BIP * (team (H-HR)/BIP). Essentially, the Voros McCracken number. For a team, Delta-H should be zero. Positive numbers signify more hits allowed than expected ("bad luck," if you believe pitchers have nothing to do with the outcome of a BIP), negative numbers mean fewer hits than expected ("good luck").

So did Sutton suddenly learn something new in 1972? Or does that simply signal the shifts in defensive lineups?

[Dark Horse]

1966 Was his rookie year. He was on a staff with Koufax, Drysdale and Claude Osteen. All solid vets.

In 1972 he was a five year vet with Osteen, John, Downing and Singer.

i would theorize one or both of the following things happening. One he got better at preventing hits and/or the pitchers around him were not as good as he was at preventing hits.

[Skipaway]

Originally posted by Dark Horse
1966 Was his rookie year. He was on a staff with Koufax, Drysdale and Claude Osteen. All solid vets.

In 1972 he was a five year vet with Osteen, John, Downing and Singer.

i would theorize one or both of the following things happening. One he got better at preventing hits and/or the pitchers around him were not as good as he was at preventing hits.

Let's just examine the differences between 1971 and 1972, where Sutton got a 28 hit jump:

Rotation:

1971
Osteen 259
Sutton 265.3
Downing 262.3
Singer 203.3
Alexander 92.3

1972
Osteen 252
Sutton 272.7
Downing 202.7
John 186.7
Singer 169.3

Defense:
1971
Sims C
Parker 1B
Lefebvre 2B
Garvey 3B
Wills SS
Davis OF
Crawford OF
Buckner OF


1972
Carnnizzaro C
Parker 1B
Lacy 2B
Garvey 3B
Russell SS
Davis OF
Mota OF
Robinson OF

Hmmm, I would believe the middle infielder change is more likely the root cause.


Koufax last pitched in 1966, and Drysdale in 1969, Osteen lasted till 1973. Tommy John is in the same rotation from 1972-1978. None of these seemed to be effecting Sutton's delta-H numbers.

[gmo]

If an adjustment was made to add variance to BABIP, what would happen? If some pitchers were made to be better than others, would it be discernible? If some small BABIP skill was introduced (or already) into the game, with all the noise in the system would the resulting performances and thus stats be useful in determining which pitchers have it and to what extent?

I took the data I posted above (a page or so back in the thread) and cut it up a little bit more. For each SP name I made 24 "careers", by for each of the 6 incarnations taking four independent 12-year chunks of data. The tables entries below show the BABIP for each "career". Also shown are the raw per start averages for hits on balls in play and total balls in play for each "career". I jgot this from the sums of those numbers divided by the number of starts, which was either 396 or 384 depending on whether the SP was top 2 in the rotation (Brown, Helling) or not (Rogers, Guzman, Oliver). The standard deviations of BABIP in the ten 12-careers clusters range from .0047 to .0065.

		Away				Home		
		H	BAP	BABIP		H	BAP	BABIP
Rick Helling	5.96	19.60	0.304		5.90	19.27	0.306
Rick Helling	5.84	19.59	0.298		5.95	20.10	0.296
Rick Helling	5.73	19.26	0.298		5.88	19.65	0.299
Rick Helling	5.85	19.57	0.299		6.02	19.16	0.314
Rick Helling	5.94	18.73	0.317		5.87	19.28	0.304
Rick Helling	5.59	18.89	0.296		5.77	19.50	0.296
Rick Helling	5.61	19.11	0.294		5.65	19.38	0.292
Rick Helling	5.98	19.70	0.303		5.76	19.44	0.296
Rick Helling	5.77	19.13	0.302		5.82	19.18	0.303
Rick Helling	5.80	19.21	0.302		5.89	19.53	0.302
Rick Helling	5.78	19.63	0.294		5.57	19.14	0.291
Rick Helling	5.84	19.46	0.300		5.83	19.18	0.304
								
Kenny Rogers	6.20	20.83	0.298		6.08	20.77	0.293
Kenny Rogers	6.14	20.55	0.299		6.13	21.02	0.291
Kenny Rogers	6.31	20.11	0.314		5.87	20.73	0.283
Kenny Rogers	6.19	20.77	0.298		6.22	21.22	0.293
Kenny Rogers	6.13	20.33	0.302		5.99	20.67	0.290
Kenny Rogers	6.04	20.31	0.297		6.08	20.49	0.297
Kenny Rogers	6.18	20.36	0.304		6.25	20.41	0.306
Kenny Rogers	6.08	20.44	0.298		5.98	20.35	0.294
Kenny Rogers	6.21	20.34	0.305		5.98	20.33	0.294
Kenny Rogers	6.19	20.67	0.300		5.95	20.71	0.287
Kenny Rogers	6.13	20.28	0.302		6.01	20.71	0.290
Kenny Rogers	6.09	20.63	0.295		6.22	20.95	0.297
								
Jose Guzman	5.82	18.72	0.311		5.45	18.77	0.290
Jose Guzman	5.65	18.82	0.300		5.48	18.57	0.295
Jose Guzman	5.57	18.53	0.300		5.55	19.07	0.291
Jose Guzman	5.72	18.66	0.306		5.74	18.74	0.306
Jose Guzman	5.53	18.82	0.294		5.62	18.99	0.296
Jose Guzman	5.48	18.60	0.295		5.55	18.93	0.293
Jose Guzman	5.65	18.17	0.311		5.50	19.18	0.287
Jose Guzman	5.68	18.82	0.302		5.60	19.11	0.293
Jose Guzman	5.65	18.68	0.302		5.48	18.78	0.292
Jose Guzman	5.59	18.68	0.299		5.50	18.82	0.292
Jose Guzman	5.55	18.68	0.297		5.76	19.16	0.301
Jose Guzman	5.44	18.58	0.293		5.57	18.80	0.296
								
Kevin Brown	6.80	21.16	0.321		6.64	21.65	0.307
Kevin Brown	6.46	21.04	0.307		6.65	21.06	0.316
Kevin Brown	6.60	20.97	0.315		6.66	21.62	0.308
Kevin Brown	6.66	21.58	0.308		6.48	21.47	0.302
Kevin Brown	6.61	20.74	0.319		6.60	21.20	0.311
Kevin Brown	6.48	21.13	0.307		6.48	21.12	0.307
Kevin Brown	6.52	20.68	0.315		6.29	20.86	0.302
Kevin Brown	6.61	21.17	0.312		6.51	21.32	0.305
Kevin Brown	6.68	21.32	0.314		6.48	21.35	0.303
Kevin Brown	6.54	21.22	0.308		6.48	21.89	0.296
Kevin Brown	6.89	21.48	0.321		6.47	21.28	0.304
Kevin Brown	6.52	21.35	0.305		6.47	21.39	0.302
								
Darren Oliver	5.94	19.64	0.303		5.70	19.20	0.297
Darren Oliver	5.98	19.07	0.313		5.93	19.26	0.308
Darren Oliver	5.92	19.12	0.310		5.61	18.88	0.297
Darren Oliver	5.88	19.64	0.299		5.82	19.83	0.293
Darren Oliver	5.92	19.15	0.309		5.65	19.28	0.293
Darren Oliver	5.78	19.17	0.301		5.69	19.51	0.292
Darren Oliver	5.92	19.68	0.301		5.76	19.97	0.288
Darren Oliver	5.72	18.63	0.307		5.57	19.01	0.293
Darren Oliver	5.81	19.57	0.297		5.77	19.28	0.299
Darren Oliver	5.89	19.68	0.299		5.69	19.16	0.297
Darren Oliver	5.78	19.18	0.301		5.98	19.13	0.313
Darren Oliver	5.81	19.04	0.305		5.78	19.46	0.297

[nestajones]

goddammit, i got somehow logged off when i tried to reply, and my reply got lost. i'll try again quick.

most of your post shows very little understanding of statistics or DIPS, and a lot of just doesn't make much sense. but i'ma try to respond to a couple things.

Originally posted by Dark Horse
i did want to make one final plea for the game to take into consideration the ability of some pitcher to “get batters out”.

assuming a starting pitcher allows around 700 balls into play, the difference between the absolute best and absolute worst player would be around 40 hits… for a whole season. A little over one hit per game started. Insignificant.

Someone responded that they thought your "insignificant" comment was tongue in cheek. Based on your following comments, I think you were serious. I disagree. A hit per game is significant.

My answer to this is the game already has added elements that are minimal. Using v6 I simmed a season and looked at the final stats. The two qualifying pitchers with the worst HR/9 were J. Axelson and D. Mizrahi. They both had 1.8 HR/9. The two qualifying pitchers with the best HR/9 were J.Burchell and M. Eck. The both had 0.2 HR/9. That is a difference on 1.6 HR per game. The difference between qualifying pitchers that gave up the most and least home runs is 38. J. Axelrod had the worst home run ratio (HR/BF) at .045; that is 38 HRs against 836 batters faced. J. Burchell had the best at .006 [6/1001]. That is a difference of .039. That’s lower than the BABIP in all but three of the seasons that Bleacher Bum researched. But in OOTP6, a pitcher’s ability to prevent HRs, even though the difference in this ability between the best and worst pitchers translates to a number that is at best slightly more significant that the ability to prevent hits.

etc
etc

What? My head is spinning. Maybe i'm just dull tonight, but your comparisons of spreads between worst and best just don't make much sense. You're comparing things which have nothing to do with each ot. Your point is lost on me.

The game is not trying to include only things which are hugely significant. It is including things which can be measured meaningfully.

If you figure a shortstop will get around 600 chances during the season. That’s 24 outs that the worst fielder would not get that the best would.

Rating SS by fielding % doesn't mean a lot. We're talking about BIP turned into outs. You need to look at assists per inning (and even that will be deceptive).

My argument is not that the way that OOTP6 uses the DIPS theory is completely wrong. In fact as i said in an earlier post, this is an improvement on v5. My point is that it is at least as important as several other parts of the game. My thought is that baseball is a game that is made up of small parts. Some of the stats that i mentioned show that. And baseball strategy is getting every advantage you can know matter how small. We have to decide whether we want to put a CF in that makes 0.5 more plays per game. Or if we want the CF that will get 10 more HRs in a season. Or a SS that who has a 1 percent less chance of making an error or the SS that will get 15 more hits in a season. Highandoutside said in this thread that “oftentimes baseball games come down to one or two important moments.” Now, it’s the bottom of the ninth, with a runner on second and two outs. Who would you rather have batting for you? A hitter that bats .300 or one that hits .242? That’s the average difference i pointed out earlier, from Bleacher Bum’s post. And, if we can make that choice with our batters, why can’t we make that decision for our pitchers? If it’s two outs and no one on, we can choose a pitcher who’s HR/9 is no more than 1 HR/9 better than the next. Why can’t we make a decision on a pitcher who is no more than 1 hit per game (with respect of BABIP) better?

Because you don't know who that pitcher is! Any minimal ability a pitcher has to affect BABIP is not clear until he has been around many many years. If a pitcher is 40 hits better than someone else one year, that's not likely to happen the next season.

have you been reading this thread? you seem to have missed the entire point. the pitcher who happens to give up fewer hits probably doesn't have any ability at it.

[Dark Horse]

Skipaway,

My argument is not that a pitcher has total control of whether a hit occurs. It is that a pithcer has some control. This doesn't discount luck or fielding as an agent of change.

Using the example you uses from 1971 to 1972, with Russell replacing Wills at shortstop.

Sutton's change was a 28 point drop in BABIP.
Osteen's change was a 36 point drop in BABIP.
Singer's change was a 81 point drop in BABIP.
Downing's change was a 10 point increase in BABIP.

That's the thing about a pitcher's ability to prevent hits cannot be seen in a particular stat. But the ability can be seen in respect to every other pitcher on that same team. 3/4 starters experienced that same change.

In 1973 Sutton went up 37 points.
Osteen went up 28 points.
Downing dropped 14 points.


The variance around it would be the luck that is involved in the game.

If there was not an ability to prevent hits. The variance from pitcher to pitcher on the same team from year to year would be totally random.

Simply put. i expected that since Sutton's BABIP dropped 28 points that the other's starters BABIP would drop, also.

[Malleus Dei]

Spinning wheels, and dead horse beating...sigh.

[Skipaway]

Originally posted by Dark Horse


My argument is not that a pitcher has total control of whether a hit occurs. It is that a pithcer has some control. This doesn't discount luck or fielding as an agent of change.

I still think you should read through:

http://www.baseballprimer.com/artic...4-02-29_0.shtml

Think it over, then try again.