Comments on: POLL: Nolan Ryan and the Hall of Fame This and that about baseball stats. Tue, 16 Jul 2013 17:01:55 +0000 hourly 1 By: Fireworks Sun, 15 May 2011 19:01:36 +0000 @JohnAutin

I'm the one that said that early in his career that Ryan was told to reduce his wildness and didn't have much success doing it. While a comment of that nature from Ryan or people that worked with him early in his career is perhaps self-serving, and your statistics of his performance with the bases loaded is interesting, I'm not particularly moved. If all other things were equal, then you'd have a case, but the myriad of differences between most situations and a bases loaded situation don't make for a reasonable comparison. Are hitters more aggressive? I imagine they are but I don't have those figures in front of me. Are pitchers more inclined to try to get the ball over the plate, but also really wary of giving up a big knock? I imagine they are but again I don't have those figures. But that's only part of the question. The other part of the question is how does facing Nolan Ryan with the bases loaded affect your mindset/approach as a hitter? Do you tell yourself that whether he walked (or HBP) one or all of the runners on base he isn't going to want to walk you? Do you try to draw a walk and get K'd instead as Ryan trying to get it over the plate instead of being his own worst enemy yet again? What goes on in Ryan's mind? I don't think that can be answered and again feel uncompelled by your usually compelling analysis.

Additionally, I should note that Ryan was set to quit baseball very early in his career, IIRC a combination of homesickness/dislike of the big city and (probably) frustration with his performance. Considering that and that there was somewhat more bias against pitchers walking batters at the beginning of his career, I don't buy that he just 'chose' to be unhittable rather tha be more effective.

Lastly, I didn't make that comment to pretend like he was a better pitcher than he was--I made it because Ryan's critics have always pointed to his wildness, and I just don't buy that he was selfish, apathetic, or ignorant. I still like the anecdote that Randy Johnson attributes Nolan Ryan with helping him. Ryan wasn't a pitching coach, wasn't on the same team, but he saw something that no one else saw (or something that Randy had been told about but didn't listen to when received from other sources). I'm not sure how that anecdote fit into my post before but I meant for it to be as evidence of his maturity as a pitcher over the years.

By: Rich Sat, 14 May 2011 21:27:18 +0000 Dave V, you may be severely underestimating what a crazy pitcher's park the Astrodome was, esp at that time.

By: Timmy Patrick Sat, 14 May 2011 10:37:16 +0000 @186 Dave V. - Great example, I am always skeptical of WAR. I don't mind so much if it's comparing the best of the best, but if it's used to keep a player out of the HOF I don't like it.

By: Jared Fri, 13 May 2011 19:14:22 +0000 A few things to consider for those discussing ballpark factors between Ryan's and Niekro's 1983. It's easy to compare performance by removing the ballpark factor. This can be done by comparing each pictcher's performance on the road, meaning they each pitch in a variety of ballparks, which should even things out statistically. Ryan still outpitched Niekro pretty convincingly in virtually every category. Now, that doesn't factor in the defensive prowess of the team playing behind them. However, over the course of the season, 8 hitters reached on an error while Ryan was pitching and 9 reached on an error while Niekro was pitching. Not much difference there on this particular measure of fielding. One more thing to consider is how well Ryan pitched in Niekro's ballpark (a hitter's park) and how well Niekro pitched in Ryan's ballpark (a pitcher's park). Looking at the splits, Ryan pitched in Atlanta once that year, allowing 2 hits and 2 runs over 8 innings. Niekro pitched once in Houston, allowing 8 hits and 3 runs in 4.2 innings. You can't speculate too much over one start each, but it highlights the problems with applying adjustments based on the aggregate to individuals. It puts you in a position where you are adjusting individual stats due to aggregate base rates, which always introduces error. One last note, ERA+ is adjusted to the ballpark, and Ryan's was 114 and Niekro's was 98, pretty much nullifying the ballpark factor. WAR can be a useful number, but I think the comparison of the two seasons shows that it has flaws just like any other advanced stat.

By: Dave V. Fri, 13 May 2011 16:22:04 +0000 I don't dismiss anything that conflicts with the raw numbers and I tried to convey that when mentioning "I know that one example of stats doesn't invalidate an entire system" (though I posted that sentence incorrectly in my first post and made the correction in the post right below) 🙂 I do however think that WAR is wrong in this specific case for sure.

Niekro had two teammates in 1983 who had pretty successful seasons despite pitching in the same ballpark and having the same defense that Niekro did. Craig McMurty and Pascual Perez finished 10th and 17th respectively in pitcher WAR that season. And for the most part, both seem to have respectable raw numbers unlike Niekro. They managed to pitch well, as did Ryan (they all dealt with the hand they were dealt). Niekro didn't pitch well, at least this season IMO. He shouldn't get undue credit just because of where he pitched. Sometimes a pitcher in a tough pitching situation is just a mediocre or bad pitcher IMO (obviously I'm not saying Niekro was mediocre/bad his whole career of course).

By: Johnny Twisto Fri, 13 May 2011 15:51:06 +0000 Dave, there's nothing wrong with looking at the raw numbers. They are what actually happened. But if you dismiss anything that conflicts with them, then you should ignore WAR. There would be no point in having WAR if it just rubber-stamped what you already know to be true. I mean, this is kind of the goal of sabermetrics in general: the search for *objective* knowledge about baseball. Ryan pitched in a more favorable park, and probably had a superior defense, and yes, each had to deal with the hand he was dealt. WAR attempts to strip away those contextual factors which were outside the pitcher's control. If you think a 3.00 ERA is always better than a 4.00 ERA, that is your right. I believe sometimes the 4.00 ERA has more value. Context context context.

By: Michael E Sullivan Fri, 13 May 2011 15:50:29 +0000 You are comparing a pitcher in one of the best hitters parks to a pitcher in one of the best pitcher's parks. Raw numbers are going to be worthless. It's like looking at the raw stats of a hitter in coors in 2000.

If you compare raw numbers of pitchers in comparable parks, then yes, you'd think if pitchers were comparable, then most of the raw numbers would be comparable, or favoring different guys.

With felix vs. CC, the raw numbers (except for W-L where being on the Yankees vs. Mariners was a huge advantage for CC) actually favored Hernandez unfairly, as he was in a real pitcher's park, while CC was pitching in a hitter's park. Looking purely at raw numbers (and ignoring w-l) the gap looks much larger than it really was between felix and CC. Felix is still comfortably ahead after making adjustments for these things, but it would not look as dramatic in the raw stats if their parks were not so disparate.

Raw numbers are nearly worthless for making fine distinctions. If a guy has the best raw numbers in the league, sure, you can tell he's at least really good, and if he has the worst, he's probably really bad.

But for deciding between two guys who are both pretty good, both mediocre, or both excellent, or deciding whether somebody is pretty good or excellent, or a bit below versus a bit above average, or a bit below versus replacement level -- raw numbers are useless.

By: Dave V. Fri, 13 May 2011 15:28:57 +0000 @189 - Ryan gave up 9 unearned runs and Niekro gave up 5 unearned runs. If you look at total RA, Ryan moves from a 2.98 ERA over to a 3.40 RA. Niekro moves from a 3.97 ERA over to a 4.20 RA. Still a big edge for Ryan.

I get that WAR accounts for park factors and defensive metrics as mentioned and you mention "looking at raw numbers will lead you very very wrong here" but IMO, when you can't look at raw numbers at all, then that's a problem.

I mean, Ryan beats Niekro in every single raw number. If there is literally nothing at all visible to the naked eye to show that Player X is better than Player Y and only a calculator can show that Player X is equal (or better than) than Player Y, well that's not good.

Take Felix Hernandez and his Cy Young last year. Though he didn't have a gaudy W/L record, you could still look at his raw stats and see that he had a great season. You could directly see that he outpitched everyone else in the AL (and thus was a deserving Cy Young winner). Now obviously neither Ryan nor Niekro were Cy Young-caliber in 1983. But when we get to a point where actual raw stats can't be looked at to tell a picture, I'm kind of at a loss. Some pitchers have good seasons despite pitching in a hitter's park and/or having a weak defense. Some pitchers have bad seasons despite pitching in a pitcher's park and having a great defense. You deal from the hand you're dealt. Saying that a guy who pitched well is equal to a guy who pitched poorly is wrong, regardless of the circumstances IMO. I'm a fan that tries to look at things from both an "old-school" and sabermetric viewpoint, but if this makes me sound "old-school" (and thus in some people's eyes, uninformed), well so be it.

By: Michael E Sullivan Fri, 13 May 2011 14:54:17 +0000 "I don't have enough info to state whether Ryan should have had a higher WAR relative to all the other pitchers in MLB that year. But there's no way Niekro should be as high as Ryan. Park factors,m defensive metrics and whatever else you want to account for, there's just no way Niekro pitched even close to Nolan Ryan in 1983. "

First, looking at raw numbers will lead you very very wrong here, because Niekro pitched in a major batters park, while Ryan was pitching in a major pitchers park (108 v. 93). So looking at WHIP, H/9, BB/9, K/9 and raw ERA are going to make Ryan look *much* better than he really was relative to neikro.

So the huge ERA gap, when corrected for park is only *1/2* a run, rather than a full run.

Next, WAR looks at all runs, not merely earned runs, and Niekro gave up fewer unearned runs than did Ryan.

Secondly, WAR adjusts for fielding. The astros in 1983 had an excellent fielding team worth 34 runs over average according to TZ. The braves had a relatively poor fielding team worth -16 runs on average. That's a difference of 50 runs over the season which works out to about 7 runs in 200 IP.

So Ryan saw 74 runs with 6.85 saved by his fielders (relative to Niekro), while Niekro saw 94 runs. 81*(1.08/.93) for the park adjustment is 93.7 runs. Given that Niekro pitched 5 more innings, by this metric they were just about the same.

Since that is the metric WAR uses, that is the reason they come out to having the same WAR.

If you don't like using total runs, you can do the calculation using ER only. Then you get 89 runs for niekro, and adjusting you get 83.43 for Ryan. That will give a WAR difference of around .56

The real answer to who pitched better is probably somewhere in that range. It's likely that Ryan did have a slightly better year and that some of the difference in unearned runs was due to luck rather than skill. So I suspect that WAR is off here, but not by nearly as much as you seem to think.

By: Michael E Sullivan Fri, 13 May 2011 14:23:41 +0000 Timmy. I did not say coriolis doesn't affect outfield throws. I said it affects all outfield throwers the same way. They all have to aim something like a few millimeters off (or less depending on how fast they throw, and the compass orientation of the throw). Doesn't matter whether they are righty or lefty. The ball will end up in a slightly different spot than if there were no coriolis. But it's the same for everybody, and in adjusting for the park and the breeze, they will automatically also adjust for coriolis.

The only way it affects left or right handers differently is if you are trying to make the ball curve sharply left or right. Then, in certain orientations and certain kinds of curves, it can work for or against your natural curve. The only players who care about the break of a ball rather than just where it ends up are pitchers, so that is the relevant calculation.