Sunday, May 27, 2007

My Top 60 Starters, Warmup

In the next installment I will get around to unveiling (a nice word to build suspense and anticipation where none should exist) spots 51-60 on my list; for now, I am going to discuss some of the pitchers who did not make the list and how the average and median Hall of Famer does.

Here are the figures for the average and median Hall of Famers:
CAT….NW%...WAT.…….WCR…...ARA……..WAA………....WAR………......Top 5

To me, anyone who is better then the median Hall of Famer should probably be there himself, no questions asked (all other things being equal). It is those below the median for whom there is a real debate. There are 31 pitchers with more career WAR then the median Hall of Famer, and all who are eligible are in the Hall with the exception of Bert Blyleven.

Now, let me tell you that pitchers 61-73 on my list, in rough chronological order are: Sam Leever, Ed Reulbach, Chief Bender, Carl Mays, Bob Shawkey, Dolf Luque, Burleigh Grimes, Herb Pennock, Bucky Walters, Jerry Koosman, Jim Kaat, Bret Saberhagen, and David Cone. To be honest, by the time I got to around 45 it was very difficult to make judgments on specific pitcher-to-pitcher comparisons; there are too many guys with similar WAR, similar WAA, similar ARA, etc.

Among pitchers who didn’t make it at all, there are few that I will discuss. The first is Eddie Cicotte. The numbers alone would have ranked Cicotte in the mid-40s, but I docked him the entire 1919 season, which was his second-best season, +11.8 WAR. That drops him from the mid-40s into the range of borderline top sixty, and I chose other pitchers ahead of him. If you throw the World Series, then you have in my eyes destroyed your body of work in the regular season. Add in the fact that his fourth-best season, 1920 (+7.6 WAR), has to be viewed with some suspicion given the fact that he had already knowingly thrown games at the time, and it’s easy to drop him past the very similar numbers of others in that group. On talent alone, Cicotte would belong.

Jack Morris is a pitcher whose reputation is just not matched by his performance. The reason he has been overrated is that he does well in W-L record; his +74 WCR ranks thirty-eighth all-time. But his career ARA was just 105; he was only +63 WAR for his career. He COULD be ranked in the top sixty, particularly if you want to give more weight to the W-L, but I can’t justify it myself.

Mel Harder is a guy who still gets Hall of Fame push, but maybe that is only my impression because I am from Cleveland and read old newspaper columnists wax poetic about the olden days of good old Mel. After all, the famous Bill James Keltner List was the result of old Keltner fans pushing him, and was discussed in the Indians essay of one of a Abstract. Mel Harder was a fine pitcher, no doubt; .534 NW%, 108 ARA, +58 WCR, +61 WAR. But what sets him apart from Milt Pappas (.550, 111, +60, +61)? Or Freddie Fitzsimmons (.572, 109, +66, +59)? Or Orel Hershiser (.567, 109, +63, +59)? You get the idea. There are too many pitchers with near identical characteristics. These are the guys who would make up the bottom portion of the Top 100. Excellent pitchers, not Hall of Famers.

Well, that is, except for the Hall of Famers who don’t make my top sixty. Leaving out Bender and Grimes, who were in my 61-70 range, they are: Lefty Gomez, Catfish Hunter, Bob Lemon, Jesse Haines, Jack Chesbro, Dizzy Dean, and Rube Marquard.

Gomez pitched for great Yankees teams, and ranks much better versus average then he does against replacement. Jimmy Key has very similar numbers, though, with the exception of Gomez’ higher peak. Nobody thinks of Key as an all-time great.

The late Catfish was by all accounts a great guy, and he pitched for six pennant winners. But his career was short, and if you compare him carefully to his teammate Vida Blue, it’s tough to pick one. Blue is viewed as a disappointment because of his brilliant early work, followed by fairly average pitching, but:
Hunter did that in 3449 IP, Blue in 3343.

Bob Lemon had a short career (2850 IP). His teams had excellent records (.589, about the same as the .587 for Whitey Ford). He was a converted third baseman, so he wasn’t a bad hitter, but +61 WAR needs a lot of help to get into HOF territory. He could get some war credit, but he missed his age 22-24 seasons and was not a pitcher before he left, so that’s extremely iffy to me.

Jesse Haines was one of the Frankie Frisch crony choices, and has absolutely nothing to set him apart from the pitchers in the Mel Harder pack, as he was .549, 108, +59, +58 himself.

Jack Chesbro is probably in the HOF because of his tremendous 41-12 season in 1904. That was a legitamitely great season (+14.8 WAR), but he never again cracked double digits, so his peak is not THAT spectacular (+45 WAR in top 5 years ranks thirtieth). But if you focus on peak value, he has a case to be one of the greats.

Dizzy Dean was brilliant as well, but only for a brief time. His top 5 seasons are +42, good enough to rank 39th, so he’s a peak special. His rate stats are great (.625 NW%, 131 ARA), but he was 33 innings short of 2000. Just not enough career value for me.

Finally, Rube Marquard’s inclusion in the Hall of Fame is a joke. Marquard was slightly above average on a rate basis (.517 NW%, 104 ARA), and didn’t have a super-long career in order to provide that much value (+53 WAR). Compare him to Claude Passeau, who I doubt many have even heard of (.512 NW%, 110 ARA, +52 WAR). Marquard is, without any doubt in my mind, the worst pitcher in the Hall of Fame. He wouldn’t make my top 120.

Then there are the pitchers who are included in the Hall of Merit but not in my top sixty. These I will treat in more detail then the HOF snubs, because I think the HOM voters are much more qualified to do this task, and since there are not the personal politics of the Frisch-type that got us Jesse Haines.

Alright, I lied, there’s no “pitchers”. There’s a pitcher, singular, and that is Wes Ferrell. Let me start by giving you my evaluation of Ferrell, and then let’s look into arguments put forward in his favor. Luckily, Dick Thompson does not read my blog or know me from Adam, so we can do this rationally.

Ferrell has a great W-L record, there’s no doubt about it: .607 NW%, +35 WAT, +70 WCR (45th in WCR). However, his ARA is only 112, for +15 WAA and +52 WAR. He pitched only 2630 innings, and won just 18 games after the age of 30.

One pro-Ferrell point is that he had a fine peak, which is true, but given the ground rules here, is irrelevant. Another is that he was an excellent hitter. This we can quantify, so let’s get at it. Ferrell 1305 career PA, hitting 280/351/446. This is excellent, but not quite as good as it looks, as he played in a high-scoring context (N=5.22, PF=1.02). His RG of 5.60 is +13 runs versus an average hitter.

Of course, the standard should not be an average hitter, but an average hitting pitcher. The average pitcher hit at 40% of the league RG in the 1930s (a .138 OW%). Ferrell is +118 runs against this standard.

But in fact this is too much, because Ferrell should only be compared to a pitcher when he is actually pitching. Ferrell played in a total of 547 major league games, but only 373 of these were as a pitcher. That means that in 174 of those games, he needs to be compared to a replacement level hitter, not an average pitcher.

Unfortunately, we don’t know how many PA he had in those games, but let’s be conservative and assume it was just one per game. Sure, there may have been games in which he pinch ran or something and never hit, but there were also likely games in which he had multiple PAs. Ferrell played in 13 games in the outfield for the Indians in 1933, his only games in the field, and recorded 2.46 Total Chances/Game, versus a league average of 2.28, so it’s safe to say he was playing the majority of those games.

So we assume that Ferrell had 1131 PAs that need to be evaluated v. an average pitcher, and 174 that need to be evaluated v. a replacement hitter (73% of the league average, .350 OW%). Under this new standard, he is +110 runs, which translates to +10.5 wins.

These are extra wins that we can add in to his pitching performance, and so instead of being +52 WAR and +15 WAA, he can be +64 WAR and +26 WAA. This would definitely bring him closer to the top 60, however there are twelve pitchers with the higher or same WAR not in, and many others right behind them. The WAA fares better; there are only six in the neighborhood. So perhaps I should have put Ferrell in the 61-70 range, but he still doesn’t crack the top sixty.

Some of the other arguments for Ferrell centered around peak value by comparing him to Grove, which is not something I’m going to explore here because peak is not on the table at all. Another was that Ferrell had to pitch more against the better teams in the league because he was his team’s ace, and a poor team needed to throw their big gun to have a shot against the Yankees and the other contenders.

“Jonesy” provided in-depth data that he researched for 1932, which I rate as Ferrell’s fourth-best season (+6.8 WAR). Here are Ferrell’s IP v. each team, along with their RG:
NY: 42.1 IP, 6.42 RG
PHA: 41 IP, 6.37 RG
WAS: 44 IP, 5.45 RG
DET: 47.1 IP, 5.22 RG
SLA: 33 IP, 4.77 RG
CHA: 52.1 IP, 4.39 RG
BOS: 27.1 IP, 3.68 RG
Weighting each team’s RG by the percentage of his IP thrown against them, the true context Ferrell pitched against was 5.25 runs/game. The assumed context was 5.33 r/g, so this in fact hurts Ferrell, ever so slightly. The point of this data was supposed to be a comparison with Grove, but that is irrelevant here; the conclusion is that in 1932, at least, Ferrell’s assumed and actual contexts were essentially equal, and no adjustment is needed.

I don’t know if other data was compiled broken down by season, but if so I have not found it. The bottom line is that Ferrell is a borderline top 70 pitcher by my standards. He is not a HOFer by these standards.


  1. Brandon,

    How could you have missed Chris Jaffe's spectacular series on pitcher leveragring:

  2. Guilty as charged. If it makes it any better, I wrote this series way back in Jan or Feb and am just getting around to posting it now. But I certainly a) should have read Jaffe's articles and b) acknowledged his research.


Comments are moderated, so there will be a lag between your post and it actually appearing. I reserve the right to reject any comment for any reason.