The Editor's Disclaimer HerEx intro Hardball home

Angels' Davis finding AL to his liking

By Kenneth Broder
Los Angeles Herald Examiner
July 17, 1988

The Angels' Chili Davis reportedly has a $100 bet with Joe Carter of the Cleveland Indians over who is going to have a better year in 1988. Actually, this is the double-or-noting follow-up to last year's bet which Chili lost decisively enough to drive him out of San Francisco.

The Angels' announcers who were discussing this bet in a game recently acknowledged that Carter had the edge again this year, but thought that Chili was still in the hunt. After all, both players were hitting .284 at the All-Star break, though Carter led in homers (16-12) and RBIs (63-54). The difference between the two players appeared to be just a couple of big games for Chili.

Those stats, however, are a bit misleading. We could point out that Carter has stolen 10 more bases than Davis, while being caught one less time. Or that Carter has hit an extra triple, rapped four more doubles and walked as much as Davis -- all while making 11 fewer outs.

But none of this makes an overwhelming argument against Chili Davis, until you pull all these disparate numbers together with the Runs Created formula and discover that a lineup of nine Chili Davises would theoretically score 5.26 runs per game compared to 6.12 for a lineup of Carters. That's about a 16 percent difference. You give the Angels 16 percent more runs and they would be leading the American League in scoring by 12 runs instead of ranking eighth.

At this pace, Carter alone will be responsible for producing about 20 more runs than Davis. But I seriously doubt that either one of them has heard of Runs Created, or any of the other sundry ways of measuring offense and will be haggling at the end of the season over the relative value of Chili's batting average versus Carter's RBI total.

Davis might not quite measure up to the standard being set by Carter, but he's one of the few Angels regulars having a better-than-usual year. Going into this season, Davis had a Runs Created Per Game average of 4.91; this year he's improved that by 7 percent. The improvement is even more marked when you consider that the average team in the American League scores about 4.4 runs per game, down almost half a run from last year.

Davis has upgraded his game by becoming more aggressive at the plate. His batting average is up, as well as his isolated power (slugging percentage minus batting average), while his walks per plate appearance have plummeted from a moderate 1-per-9.72-plate appearance level to a wild-and-woolly 1-per-15.39 mark.

Only Johnny Ray, coincidentally another castoff from the National League, has shown a similar improvement on the Angels. (See chart.) A lineup of nine Rays would theoretically score 5.44 runs per game this year compared to a career mark of 4.70. Also like Davis, he's done it by boosting his batting average and isolated power while walking less.

(Hmmm. Is there a potential study here of what happens to the offense of ballplayers who, in their primes, suddenly begin walking less? Or more?)

So far this year, Ray and Davis have been the most productive hitters on the team, though it remains to be seen how many runs Davis has cost the team with his glove. There are a few complicated formulas out there for taking a stab at figuring that out, but I'm real skeptical of fielding stats. They tend to prove beyond a doubt that Buddy Bell as a better thirdbaseman that Brooks Robinson and Glenn Wilson deserves to have his glove bronzed and in the Hall of Fame.

Ray and Davis are not only above average in production per at-bat, they are playing regularly. It doesn't do much good to have a great Runs Created Per Game average and not play many games. By combining playing time and RC Per Game you can treat a ballplayer's offensive record like a team's won-loss record and figure out how many more wins that losses a player contributes to his team. Just think of a hit or walk as contributing to a win and an out as contributing to a loss.

If you have a weak stomach or a slight headache, skip this paragraph. Otherwise ... the formula works like this: 1) Square a player's RC Per Game; 2) Square the number of runs scored by the average team per game; 3) Add those two numbers together; 4) Divide the original RC Per Game squared by the number derived in Step 3 and you have a player's offensive won-loss percentage; 5) Divide the number of outs a player made by 25.5 (approximately the number of outs in a game) to represent the player's Games Total; 6) Multiply the figure from Step 4 by the player's offensive won-loss percentage from Step 5 and you have the number of wins he contributed to his team; 7) Subtract those wins from the player's Games Total in Step 5 and you have his losses.

Despite the off years suffered by many of the Angels hitters, the team has been doing something right. The Runs Created formula predicts the team should have scored 349 runs; they've actually scored 383. That could be attributable to clutch hitting, good managing ... or maybe dumb luck.

Whatever the reason, the hitters' good fortune hasn't rubbed off on the team's won-loss record. When you apply the aforementioned Pythagorean technique to the Angels as a team, using their runs scored and opposition runs as elements, you get a projected won-loss record of 41-48, precisely their record at the All-Star break.

No team's record in the American League so accurately reflects its runs/runs allowed performance. If the Johnson Effect holds true -- and teams that outperform their predictions for a while tend to underperform at some point -- look for Toronto, Boston, Milwaukee and New York to find their fortunes turning upward while Cleveland, Texas and especially Chicago suffer a run of "bad luck."


2.Davis C.


9.Davis D.




* RC is Runs Created. RC = (Hits + Walks - Caught Stealing) x (Total Bases + .55 x Stolen Bases) / (AtBats + Walks)
* RC/G is Runs Created Per Game. RC x 25.5 / (Hits + Caught Stealing)
* Pct = RC/G / (RC/G + League RC/G)
* Wins = Pct x Assigned Games (Sorry you've got to read the column for this one.)
* Losses = Assigned Games - Wins
* Diff = Wins - Losses
* Statistics are for games as of July 10. Discrepencies between Diff and Won-Loss stats are due to rounding off.