Turn Off Ads?
Page 6 of 9 FirstFirst ... 23456789 LastLast
Results 76 to 90 of 123

Thread: The Value of our 2007 Cincinnati Reds

  1. #76
    Posting in Dynarama M2's Avatar
    Join Date
    Sep 2000
    Location
    Boston
    Posts
    28,448

    Re: The Value of our 2007 Cincinnati Reds

    Quote Originally Posted by jojo View Post
    Offensive metrics like RC for instance aren't real runs. In fact there are some serious issues with their ability to capture reality.
    Really? Golly, they really ought to tell people that. I mean who knew?

    Quote Originally Posted by jojo View Post
    Of course it needs an asterisk. I don't think many would use a 23 game sample as definitive proof of anything. Does anyone think Votto is going to OPS 1.025 against righties for the rest of his career? Sample size is a consideration with defensive stats just as it is with offensive metrics.
    Except RC is telling you how many runs he created (and it correlates quite nicely with runs scored at the macro level). Feel free to tweak the RC formula, but no one's really questioning whether Votto created 17 runs of offense in his small sample size. It may have some fluke elements to it, but I've yet to see someone doubt whether it happened.

    Meanwhile, it's an open question as to whether he really cost his team 4.1 runs below an average fielder on defense. In fact, the small sample size should make that more difficult because he didn't have a lot of time to do that much extra damage. Did that really happen?

    Maybe it did, but if so that's some catastrophic defense.
    Baseball isn't a magic trick ... it doesn't get spoiled if you figure out how it works. - gonelong

    I'm witchcrafting everybody.

  2. Turn Off Ads?
  3. #77
    Five Tool Fool jojo's Avatar
    Join Date
    Nov 2006
    Posts
    19,135

    Re: The Value of our 2007 Cincinnati Reds

    Quote Originally Posted by M2 View Post
    Except RC is telling you how many runs he created (and it correlates quite nicely with runs scored at the macro level).

    Feel free to tweak the RC formula, but no one's really questioning whether Votto created 17 runs of offense in his small sample size. It may have some fluke elements to it, but I've yet to see someone doubt whether it happened.
    Actually yes it it very debatable whether Votto created 17 runs. His high SLG and OBP coupled with the small number of at bats makes it very likely that we haven't captured Votto's true performance level using RC as a metric.

    RC as an offensive metric is progressively being marginalized and really is hanging around right now because it's very easily accessible. The only reason I still use it in this community is because enough arguments get started with RC that using better metrics that haven't become mainstream to as many people would dramatically choke otherwise fun discussions.

    Quote Originally Posted by M2 View Post
    Meanwhile, it's an open question as to whether he really cost his team 4.1 runs below an average fielder on defense. In fact, the small sample size should make that more difficult because he didn't have a lot of time to do that much extra damage. Did that really happen?

    Maybe it did, but if so that's some catastrophic defense.
    Of course it's an open question.

    Maybe I misunderstood your post but it seems to be suggesting that sample size isn't an issue on the offensive side when evaluating Votto's true performance level but it makes evaluating his defense untenable.

    Votto's September isn't sufficient to make any strong conclusions about either his glove OR bat.
    Last edited by jojo; 11-12-2007 at 09:07 AM.
    "This isnít stats vs scouts - this is stats and scouts working together, building an organization that blends the best of both worlds. This is the blueprint for how a baseball organization should be run. And, whether the baseball men of the 20th century like it or not, this is where baseball is going."---Dave Cameron, U.S.S. Mariner

  4. #78
    Posting in Dynarama M2's Avatar
    Join Date
    Sep 2000
    Location
    Boston
    Posts
    28,448

    Re: The Value of our 2007 Cincinnati Reds

    Quote Originally Posted by jojo View Post
    Votto's September isn't sufficient to make any strong conclusions about either his glove OR bat.
    I'm not making conclusions about either.

    Analytics 101 - separate what did happen from what might happen in the future.

    So back to the question at hand, what was the value of his offense and his defense? The sample size doesn't matter since we're not projecting anything. We're only dealing with what Votto did in that small sample size. The problem with the defensive systems being used here is it's exceptionally hard to be 4.1 runs worse than average over 21 games of fielding time. Heck, if you look at the raw numbers, Votto missed a total of 12 balls supposedly in his zone (according to RZR) for the season (8 at 1B, 4 in LF). A good fielder would have only been expected to make an extra 5 plays at 1B and 2 in LF. So 4.1 runs below average on what, if we stretch it, is 7 plays seems like a major stretch.
    Last edited by M2; 11-12-2007 at 11:42 AM.
    Baseball isn't a magic trick ... it doesn't get spoiled if you figure out how it works. - gonelong

    I'm witchcrafting everybody.

  5. #79
    Five Tool Fool jojo's Avatar
    Join Date
    Nov 2006
    Posts
    19,135

    Re: The Value of our 2007 Cincinnati Reds

    Quote Originally Posted by M2 View Post
    I'm not making conclusions about either.

    Analytics 101 - separate what did happen from what might happen in the future.

    So back to the question at hand, what was the value of his offense and his defense? The sample size doesn't matter since we're not projecting anything. We're only dealing with what Votto did in that small sample size. The problem with the defensive systems being used here is it's exceptionally hard to be 4.1 runs worse than average over 21 games of fielding time. Heck, if you look at the raw numbers, Votto missed a total of 12 balls supposedly in his zone (according to RZR) for the season (8 at 1B, 4 in LF). A good fielder would have only been expected to make an extra 5 plays at 1B and 2 in LF. So 4.1 runs below average on what, if we stretch it, is 7 plays seems like a major stretch.
    The simple answer is the value of his defense was -4.1 using the average of Justin's methodology of combining data from Fans' Scouting Report, ZR, RZR and then applying positional adjustments. The value of his offense was 7 RAR according to Justin's methodology.

    I'd suggest it's pretty easy to perform at an extreme relative to the mean over a smaller sample. I don't have a big issue with the notion than 7 plays could translate into something approaching 4 runs. Balls down the line/in the gap by definition are likely to lead to higher run scoring because they'll most likely be XBH.

    Here's some "back of the napkin scratch" to illustrate the point by looking at the issue from the perspective of the offense. You suggested that the "good fielder" would have been expected to convert 28 out of Votto's 33 chances so the offense would hit 5 for 33 in that scenario. If we assume two of those hits were doubles, the hypothetical offensive player would have a line of .152/.152/.212. The hypothetical offensive player in Votto's circumstance would be 12 for 33 (since RZR suggested he missed 7 more plays than what you thought a good fielder might be expected to make). If we conservatively assume that just 2 of those additional hits went for XBH in the form of doubles, the hypothetical offensive player in Votto's case would have a line of .364/.364/.484. Since you favor what Bill James has to say about OBP and SLG in relation to offensive value, over those 33 ABs, the good fielder's offensive counterpart would've created 1.1 runs while Votto's counterpart would've created 5.8 using the simple RC formula.

    BTW, Bill James suggests a rough estimate method for converting Dewan's +/- totals into runs is to simply halve the totals. So Bill James would suggest 7 plays would equal something like 3.5 runs. Dewan since seems to be favoring little higher mutliplier (times #plays by .65).
    "This isnít stats vs scouts - this is stats and scouts working together, building an organization that blends the best of both worlds. This is the blueprint for how a baseball organization should be run. And, whether the baseball men of the 20th century like it or not, this is where baseball is going."---Dave Cameron, U.S.S. Mariner

  6. #80
    Stat Wanker Hodiernus RedsManRick's Avatar
    Join Date
    Dec 2004
    Location
    Guelph, ON
    Posts
    16,181

    Re: The Value of our 2007 Cincinnati Reds

    It seems like the crux of much of the debate isn't so much the measurement of what happened in terms of plays made or not made, but the valuation of those plays not made.

    If we're taking 7 should have been outs and turning them in to hits, 4 runs in 7 plays is .57 runs per play. That doesn't exactly strike me as a stretch at all.

    Per Tango, here are the average (base/out) run values for events 1999-2002
    http://www.tangotiger.net/RE9902event.html

    Code:
    		HR 	3B 	2B 	1B 	RBOE 	HBP 	NIBB 	IBB	Bunt 	Out 	K 	
    Average 	1.409 	1.063 	0.764 	0.474 	0.546 	0.385 	0.33 	0.102	-0.001 	-0.299	-0.31 	
    
    		PB 	WP 	BK 	SB 	CS	PO 	POE 	OA 	INT 	FoulE 	DefInd
    Average		0.285 	0.284 	0.237 	0.195 	-0.456	-0.255 	-0.953 	-0.572 	0.429 	0.026 	0.063
    Do those figures strike you as particularly off M2? Or do you believe it's not likely that Votto really missed that many plays? I'd certainly like to see the plays he missed, as 7 in 24 games seems like a lot to me, but equating 7 missed plays to 4 runs doesn't seem odd.
    Last edited by RedsManRick; 11-12-2007 at 01:15 PM.
    Games are won on run differential -- scoring more than your opponent. Runs are runs, scored or prevented they all count the same. Worry about scoring more and allowing fewer, not which positions contribute to which side of the equation or how "consistent" you are at your current level of performance.

  7. #81
    Member SteelSD's Avatar
    Join Date
    Mar 2002
    Posts
    9,391

    Re: The Value of our 2007 Cincinnati Reds

    Quote Originally Posted by jojo View Post
    BTW, Bill James suggests a rough estimate method for converting Dewan's +/- totals into runs is to simply halve the totals. So Bill James would suggest 7 plays would equal something like 3.5 runs. Dewan since seems to be favoring little higher mutliplier (times #plays by .65).
    Have you questioned Dewan's multiplier? Are we really to believe that nearly two-thirds of "missed" plays come around to score?

    In 2007, MLB teams combined for 18,365 Runs, excluding Home Runs (which cannot be "missed" plays). In total, we saw 36,224 non-HR events (non-HR Hits + BB + HBP + CI). That sets the bar at @50%. Add in BIP Errors that allow runners to reach base, and we likely see something in the 45-47% range. I'm not sure that Dewan's +/- system is accurate to begin with, but if he's using a .65 multiplier, I'd suggest he's off his rocker even if his system is a vague representation of reality.
    "The problem with strikeouts isn't that they hurt your team, it's that they hurt your feelings..." --Rob Neyer

    "The single most important thing for a hitter is to get a good pitch to hit. A good hitter can hit a pitch thatís over the plate three times better than a great hitter with a ball in a tough spot.Ē
    --Ted Williams

  8. #82
    Stat Wanker Hodiernus RedsManRick's Avatar
    Join Date
    Dec 2004
    Location
    Guelph, ON
    Posts
    16,181

    Re: The Value of our 2007 Cincinnati Reds

    Quote Originally Posted by SteelSD View Post
    Have you questioned Dewan's multiplier? Are we really to believe that nearly two-thirds of "missed" plays come around to score?
    But do those runners put on base due to a non-play have to score to fault Votto for the additional runs, or can we also fault him for the plated runners who otherwise would have been stranded?

    The difference between an out and a single is .774 runs according to the numbers I've quote from Tango above.

    I thought we had established that the run value of an event is not just the specific likelihood of the batter to eventually score, but of the advancement of other runners already on base.
    Last edited by RedsManRick; 11-12-2007 at 01:23 PM.
    Games are won on run differential -- scoring more than your opponent. Runs are runs, scored or prevented they all count the same. Worry about scoring more and allowing fewer, not which positions contribute to which side of the equation or how "consistent" you are at your current level of performance.

  9. #83
    Posting in Dynarama M2's Avatar
    Join Date
    Sep 2000
    Location
    Boston
    Posts
    28,448

    Re: The Value of our 2007 Cincinnati Reds

    Quote Originally Posted by SteelSD View Post
    Have you questioned Dewan's multiplier? Are we really to believe that nearly two-thirds of "missed" plays come around to score?

    In 2007, MLB teams combined for 18,365 Runs, excluding Home Runs (which cannot be "missed" plays). In total, we saw 36,224 non-HR events (non-HR Hits + BB + HBP + CI). That sets the bar at @50%. Add in BIP Errors that allow runners to reach base, and we likely see something in the 45-47% range. I'm not sure that Dewan's +/- system is accurate to begin with, but if he's using a .65 multiplier, I'd suggest he's off his rocker even if his system is a vague representation of reality.
    Well, sure, if you're going to be all rational about it.
    Baseball isn't a magic trick ... it doesn't get spoiled if you figure out how it works. - gonelong

    I'm witchcrafting everybody.

  10. #84
    The Boss dougdirt's Avatar
    Join Date
    Jan 2006
    Posts
    35,933

    Re: The Value of our 2007 Cincinnati Reds

    Quote Originally Posted by SteelSD View Post
    Have you questioned Dewan's multiplier? Are we really to believe that nearly two-thirds of "missed" plays come around to score?

    In 2007, MLB teams combined for 18,365 Runs, excluding Home Runs (which cannot be "missed" plays). In total, we saw 36,224 non-HR events (non-HR Hits + BB + HBP + CI). That sets the bar at @50%. Add in BIP Errors that allow runners to reach base, and we likely see something in the 45-47% range. I'm not sure that Dewan's +/- system is accurate to begin with, but if he's using a .65 multiplier, I'd suggest he's off his rocker even if his system is a vague representation of reality.
    From what I have read in several interviews, there are different values for each types of play for each position. Its not a common 'each play is worth X value'.

  11. #85
    Posting in Dynarama M2's Avatar
    Join Date
    Sep 2000
    Location
    Boston
    Posts
    28,448

    Re: The Value of our 2007 Cincinnati Reds

    Quote Originally Posted by dougdirt View Post
    From what I have read in several interviews, there are different values for each types of play for each position. Its not a common 'each play is worth X value'.
    Well, that's just it. You've got rolling values for each type of non-out dependent on how many outs there are at the time the non-out occurs.

    But Steel's handed you the limits. Those are the total number of runs you get to work with. All the events inside that 18,365 have to roll up to 18,365. We're being asked to believe that Votto's 5-7 non-plays (and 5 is probably closer to where you'd find the average) amounted to 4.1 runs. That's an incredibly damaging set of non-plays. It would be an extraordinary combination of events. I'm not saying it couldn't or didn't happen, but right now I'm seeing no compelling evidence that it did. That's the thing about having a statistical theory, it does matter whether it translates to reality. "Well, that's what the formula says" isn't a proof.
    Baseball isn't a magic trick ... it doesn't get spoiled if you figure out how it works. - gonelong

    I'm witchcrafting everybody.

  12. #86
    Stat Wanker Hodiernus RedsManRick's Avatar
    Join Date
    Dec 2004
    Location
    Guelph, ON
    Posts
    16,181

    Re: The Value of our 2007 Cincinnati Reds

    So what are we arguing here M2? The number of missed plays or the value of those missed plays?

    If you grant the 7 missed plays (I'm not sure on what basis you are arguing the 5 number) and a non-out, non-HR play results in .47 runs on average, the math is pretty clear.

    7 missed plays x .47 runs/play = 3.29 runs. So in 21 games Votto's missed plays resulted in 3 runs, not 4. Big whoop-de-doo Let's not throw the baby out with the bathwater here.

    Whether it's 3 runs or 4 runs, clearly it's in the same ballpark. The question really is how many plays did Votto truly miss in those 21 games? Unless we have access to the PBP data, or can go back and visually evaluation the 7 they're counting, I'm not sure I see the point in debating it further.

    Can one of you clarify for me why it's ok to ignore the "not-an-out" part of the value. So if the run scoring portion of the non-play (the shouldn't have been base runner) is worth .47 runs, isn't the run prevention portion of the non-play (the should have been out) worth something too? The .47 treats the ball in play only as a part of the 36,224 event population, when the value of the play should incorporate the fact that it otherwise would have been part of the population of 130,000+ outs. We aren't going from non-existence to hit. We're going from out to hit. Or are these simply opposite sides of the same coin, not to be double counted? My thinking just isn't clear at the moment.

    I understand the importance of the distinction generally speaking, as extrapolating the difference in run valuation grows. However, it's clear in any case that poor defense can greatly cut in to the value of a player's offense and shouldn't be as discounted as it often seems to be in player valuation.
    Games are won on run differential -- scoring more than your opponent. Runs are runs, scored or prevented they all count the same. Worry about scoring more and allowing fewer, not which positions contribute to which side of the equation or how "consistent" you are at your current level of performance.

  13. #87
    Playoffs Cyclone792's Avatar
    Join Date
    May 2005
    Location
    Cincinnati, OH
    Posts
    6,284

    Re: The Value of our 2007 Cincinnati Reds

    Quote Originally Posted by RedsManRick View Post
    Can one of you clarify for me why it's ok to ignore the "not-an-out" part of the value. So if the run scoring portion of the non-play (the shouldn't have been base runner) is worth .47 runs, isn't the run prevention portion of the non-play (the should have been out) worth something too? The .47 treats the ball in play only as a part of the 36,224 event population, when the value of the play should incorporate the fact that it otherwise would have been part of the population of 130,000+ outs. We aren't going from non-existence to hit. We're going from out to hit. Or are these simply opposite sides of the same coin, not to be double counted? My thinking just isn't clear at the moment.
    Well, when a hitter hits a single, are we double counting that value too? Are we suddenly giving each offensive player .77 runs of credit for every single they hit? If we're going to double count those misplays, then that's akin to saying something like 35 percent of the total game is fielding defense. We know that's not the case so why does linear weights still try to act as if it is?

    This is the foundation of the problem with linear weights, and it's why linear weights still doesn't get it. It doesn't properly account for the relative value of offense to fielding defense, pitching to fielding defense, etc. I've said it many times before, and I'll say it again: there are only so many runs to go around, the two largest pieces of the pie aren't even going to fielding defense at all.

    Furthermore, and this is even worse for linear weights, the current era of baseball we're witnessing is one in which fielding defense is even less important than any other previous era. So not only is linear weights handing over an absurd amount of runs to fielding defense, it's doing so in an era of baseball where the value of fielding defense is probably as low as it's ever been.

    As far as Joey Votto, he didn't even appear in 21 full games worth of innings defensively, yet people seem to conclude he cost the Reds over 4 runs compared to the average defender. Stretch that out to a rate of 162 games, and that's 31 runs worse than the average as a 1B/LF. That's simply not realistic, not even close. Yes, it's a small sample size and one game could alter the results, but it's still not realistic.
    Barry Larkin - HOF, 2012

    Put an end to the Lost Decade.

  14. #88
    Please come again pedro's Avatar
    Join Date
    Mar 2002
    Location
    portland, oregon
    Posts
    14,753

    Re: The Value of our 2007 Cincinnati Reds

    Quote Originally Posted by Cyclone792 View Post
    Well, when a hitter hits a single, are we double counting that value too? Are we suddenly giving each offensive player .77 runs of credit for every single they hit? If we're going to double count those misplays, then that's akin to saying something like 35 percent of the total game is fielding defense. We know that's not the case so why does linear weights still try to act as if it is?

    This is the foundation of the problem with linear weights, and it's why linear weights still doesn't get it. It doesn't properly account for the relative value of offense to fielding defense, pitching to fielding defense, etc. I've said it many times before, and I'll say it again: there are only so many runs to go around, the two largest pieces of the pie aren't even going to fielding defense at all.

    Furthermore, and this is even worse for linear weights, the current era of baseball we're witnessing is one in which fielding defense is even less important than any other previous era. So not only is linear weights handing over an absurd amount of runs to fielding defense, it's doing so in an era of baseball where the value of fielding defense is probably as low as it's ever been.
    By that math a team that had 10 singles in a game with no other positive events (other hits, walks, or errors) would score an average of 7.7 runs a game, or 1247 runs over the course of an entire season.

    Even Ralph Wiggum knows that's unpossible.
    Get your nunchucks and the keys to your dad's car. I know where we can get a gun

  15. #89
    Stat Wanker Hodiernus RedsManRick's Avatar
    Join Date
    Dec 2004
    Location
    Guelph, ON
    Posts
    16,181

    Re: The Value of our 2007 Cincinnati Reds

    Quote Originally Posted by pedro View Post
    By that math a team that had 10 singles in a game with no other positive events (other hits, walks, or errors) would score an average of 7.7 runs a game, or 1247 runs over the course of an entire season.

    Even Ralph Wiggum knows that's unpossible.
    With Linear Weights, you have to count the values of the outs too though, right?

    So, per the table I posted earlier, those 10 singles would be worth 10*.474, but there'd also be 27 outs at -.299 for a total of -3.33 runs versus average.

    So your 10 singles would average about 1.5 runs per game.

    That said, I see Jason's point. As a system, it might work out well, but as you divy up the responsibility, those models using linear weights are giving the fielder too much credit for the outcome of a given play. We can't give the fielders full credit for the outcome, positive or negative, of those events. Sure, the RE might have been altered up to .77 runs on that non-play, however, only some portion of that can truly be attributed to the fielder.
    Games are won on run differential -- scoring more than your opponent. Runs are runs, scored or prevented they all count the same. Worry about scoring more and allowing fewer, not which positions contribute to which side of the equation or how "consistent" you are at your current level of performance.

  16. #90
    Please come again pedro's Avatar
    Join Date
    Mar 2002
    Location
    portland, oregon
    Posts
    14,753

    Re: The Value of our 2007 Cincinnati Reds

    Quote Originally Posted by RedsManRick View Post
    With Linear Weights, you have to count the values of the outs too though, right?

    So, per the table I posted earlier, those 10 singles would be worth 10*.474, but there'd also be 27 outs at -.299 for a total of -3.33 runs versus average.

    So your 10 singles would average about 1.5 runs per game.

    That said, I see Jason's point. As a system, it might work out well, but as you divy up the responsibility, those models using linear weights are giving the fielder too much credit for the outcome of a given play. We can't give the fielders full credit for the outcome, positive or negative, of those events. Sure, the RE might have been altered up to .77 runs on that non-play, however, only some portion of that can truly be attributed to the fielder.
    gotcha.
    Get your nunchucks and the keys to your dad's car. I know where we can get a gun


Turn Off Ads?

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

Board Moderators may, at their discretion and judgment, delete and/or edit any messages that violate any of the following guidelines: 1. Explicit references to alleged illegal or unlawful acts. 2. Graphic sexual descriptions. 3. Racial or ethnic slurs. 4. Use of edgy language (including masked profanity). 5. Direct personal attacks, flames, fights, trolling, baiting, name-calling, general nuisance, excessive player criticism or anything along those lines. 6. Posting spam. 7. Each person may have only one user account. It is fine to be critical here - that's what this board is for. But let's not beat a subject or a player to death, please.

Thank you, and most importantly, enjoy yourselves!


RedsZone.com is a privately owned website and is not affiliated with the Cincinnati Reds or Major League Baseball


Contact us: Boss | GIK | BCubb2003 | dabvu2498 | Gallen5862 | LexRedsFan | Plus Plus | RedlegJake | redsfan1995 | The Operator | Tommyjohn25