A baseball riddle, not to be solved

Pat, this is going to be mainly for you, but Dan, you might find it interesting in an academic sense.

Baseball is a beautiful game for a number of reasons, not least of which is the timeless debate over how games are won. We sit and witness the pitches and the hits and the plays, and yet when the whole thing is done, why did the one team beat the other? I have always contended that the starting pitcher rules between 90% and 95% of the outcome. If your starter is around league average or slightly above, and he has reasonable control, strength, and mechanics that day, your chances are pretty good. So says I.

I have seen opinions vary away from that standard; I've even seen the ratio of 50-50 pitching vs. hitting, at which I scoff! Scoff, I tell ya. And lately baseball has enjoyed an influx of geeky guys who like to take statistical samples and interpolate outcomes. This is not a bad thing, but when I have taken the trouble to explore the formulae used in some of these calculations, I'm left frowning in doubt. One of the most popular new statistical measures is WAR, or wins above replacement player. A replacement player is someone you bring in to replace an injured major-leaguer. He might have some major league experience, but isn't what you'd consider optimal. I saw recently that the standard had been calculated that if you had 25 replacement players, you'd win 43-48 games in a season. So replacement players suck, to say the least.

WAR purports to measure how many games did your team win as a result of YOU being on the team. These are games won all by your lonesome, simply as the result of your own awesomeness. It measures offensive and defensive contributions, again, purportedly. For reference, Mike Trout of the Los Angeles Angels of Anaheim had the highest WAR in the American League, 10.7. Robbie Cano of the Yankees was second, with 8.2. A 1-WAR player is estimated to be worth $4 million in salary, so, that gives you some reference.

Pitchers have WAR, too, but I struggle with that part of it. If a guy wins 15 games, or let's say his team wins 20 of the 32 games he started, how is that not, like, 20 WAR? I assume it gets down to methods of calculation. It seems like (I don't know, not being an expert in these things) they give more credit to guys who go out there every day. Okay, if your guy goes out there every day, and has, say, influence over one-ninth of the game (nine guys in the lineup, right, both at bat and in the field?), let's say over five days, he has a cumulative influence over five-ninths of the outcomes. Meanwhile, your starting pitcher had nine-tenths influence over one game. 90% is higher than 56%, isn't it?

There's another problem with a lot of these new calculations: they seem to place weight on the cumulative aspect of a lot of baseball stats. How does that help when you have a new outcome every day? The slate is wiped clean before every game, wouldn't you say?

So, now I'm going to appeal to Pat's sense of Yankee history. Baseball-reference.com is an amazingly comprehensive site. One of my extremely geeky activities is following all-time Yankee stats, and lo and behold, you can look up all-time Yankee WAR stats on Baseball-reference. Here are a couple of tables:

Rank	Player	WAR	PA
1	Babe Ruth	138.2	9,198
2	Lou Gehrig	108.5	9,663
3	Mickey Mantle	105.5	9,907
4	Joe DiMaggio	75.1	7,673
5	Derek Jeter	69.3	11,895
6	Yogi Berra	56.2	8,350
7	Bill Dickey	52.4	7,064
8	Willie Randolph	51.7	7,464
9	Alex Rodriguez	49.8	5,476
10	Bernie Williams	45.9	9,053
11	Tony Lazzeri	44.7	7,068
12	Thurman Munson	43.3	5,905
13	Roy White	43.0	7,735
14	Graig Nettles	41.0	6,248
15	Earle Combs	40.0	6,513

(PA is plate appearances)

This is a pretty cool list in that it reinforces the rough ranking that we as Yankee geeks place on our all-time guys.There are maybe a couple of surprises in there, but the overall ranking is hard to argue with. One surprise - Mantle was so much more valuable than DiMaggio, at least by this measure. Okay, let's look at Yankee pitchers:

Rank	Player	WAR	IP
1	Mariano Rivera	52.7	1,219.67
2	Whitey Ford	50.6	3,170.33
3	Andy Pettitte	45.8	2,611.00
4	Ron Guidry	45.4	2,392.00
5	Red Ruffing	41.7	3,168.67
6	Lefty Gomez	39.5	2,497.33
7	Bob Shawkey	39.0	2,490.00
8	Mel Stottlemyre	37.5	2,661.33
9	Mike Mussina	33.1	1,553.00
10	Waite Hoyt	32.0	2,273.33
11	Herb Pennock	29.9	2,202.67
12	Ray Caldwell	27.1	1,718.33
13	Jack Chesbro	26.6	1,952.00
14	Russ Ford	24.3	1,112.67
15	Dave Righetti	21.8	1,136.67

Here I get a little skeptical. Or do I? Only a few of the pitchers would make it onto the WAR chart for position players. I don't know if I agree. And how can Mariano Rivera (who has no bigger fan than me, by the way) be more valuable that Whitey Ford? Ford had greater influence over the games he pitched, right? Aren't 7 successful innings more important than one? Okay, Mariano did pitch in quite a few games, and faced a lot of high-leverage situations. Still don't know what to make of all this. There is a yin and yang of truth, and I don't know if it can be puzzled out.

I started out hoping to pose a question and a thesis about how games are won, but it's another reason the game is a beautiful conundrum: it's all right there before our eyes, day in and day out, but there is no knowing or explaining it.