Thanks, Bill, for a well-researched and written article.

There is a worthwhile discussion at Baseball Primer for those interested in whether Gibson's season was superior to those put up by Pedro Martinez in 1999 and 2000, Greg Maddux in 1994 and 1995, and Walter Johnson in 1913 (among others). The debate also discusses how much value should be ascribed to the number of "extra" innings thrown by Gibby vs. Martinez and Maddux in their best years.

Thanks for the great work, guys.

You guys should do a comparison between George Brett and Mike Schmidt. It's been somewhat-widely accepted that Schmidt is, give or take, better. But I would strongly disagree. Not only was Brett more important to his team, but his overall production was just better, too (OBP, etc). Sure, Schmidt hit twice as many home runs as Brett, but Brett did practically everything else better. Brett played longer, harder, and was just more valuable.

I'd love to get your take on this.

Bill Deane writes: "But in 1968, particularly during a two-month stretch in mid-season, Gibson was arguably the greatest pitcher of all time."

Okay, let's argue. Well, actually, I already have argued quite a bit on the SABR-list, maybe more than necessary. Email if you care to know what I said.

I am interested in what made Gibson's ERA so low that year? It was more than a run lower than his next lowest year (It would be interesting to know how many pitchers did this). Did he strikeout more batters than normal for him? Walk fewer? Fewer HRS?

His strikeout rate (strikeouts per batter faced or BFP) was .231, about 19% higher than for his entire career. But only 2% better than his next best season.

His walk rate was 36% below his career average, walking only 5.3%. That is 7.2% better than his next lowest season. But that was 1967, when he was hurt and pitched just 175 innings. The 1968 walk rate was 24% lower than the next best full season. His walk rate was about at the league average for his entire career. Yet in 1968, he had about 1 walk per 5 IP, while for his career it was about 1 for every 3 IP. The league average walk rate in 1968 was 7%. So he went well below the league average, unusual for him.

He allowed .009475 HRs per batter. That was about 59% below his career average (although his early years were in Sportmans Park which might have been a better HR park than Busch Stadium)His next best year was 1969 at .009449 and 1970 was almost as good.

His batting average on balls in play was .230, the lowest of his career (the next lowest was .240). It was .268 for his career. All Cardinal pitchers had .274 from 1960-75. So he was normally pretty close to the team average. But the team average in 1968 was .267. Some years he was higher than the team, some lower (I actually did not remove Gibson from the team calculation, so in years where he was lower than whole the team, he was even lower than the rest of the team by a few more points and vice-versa).

In 1967 he was higher than the whole team(.280 vs .270). In 1969 he was higher, .270 vs .269. I have not check every year, but 1968 is most likely the year where he went below the entire team the most.

Gibson also had allowed just a.141 average with runners in scoring position (RISP). With none on it was .180. For his career, his RISP average was .004 below his NONE ON average. Without the 1968 season, the two averages are probably very close and in general, most pitchers are a few points higher with RISP. So 1968 may have been ususual in this regard for him (although he may have had some other years where his RISP average was well below his NONE ON average and vice-versa-I will need to check Retrosheet).

Gibson was .039 lower with RISP. If I recall correctly, he had 177 RISP ABs against him that year. That amounts to about 7 hits and by some other calculations which I won't get into here, that should be about 1 run per hit. Over 305 IP, that is about .21. So we could say that he shaved that off of his ERA (so it would have still been very low, at 1.33).

But his OBP allowed with RISP was just .206, well below his NONE ON OBP of .231. Generally OBP goes up with RISP, even if you factor out intentional walks. By getting guys out with RISP, he must have prevented some rallies from taking place that would have normally made a pitcher's ERA higher.

So all of these factors combined to make his ERA so low. Which is most important, I don't know. But the low average on balls in play and low walk rate might be the biggest.

Sorry. The Cardinals allowed a .263 average on balls in play in 1968. NOt the .267 I say above.

Also, for his career, Gibson allowed a higher OBP with RISP than with NONE ON (.291 vs .299). Probably not due to intentional walks, since his IBB rate was about the same as his career rate.

Garth,

I love George Brett. However, I don't think Brett was as good as Mike Schmidt who, I believe, is the best third baseman in the history of the game.

As I summarized in an article on Wade Boggs last December, "No matter whether one prefers basic counting stats, rate stats, more advanced metrics, peak value, or career value, the conclusion is the same: Schmidt is the best third baseman of all time while Brett, Boggs, and Mathews rank second through fourth in whatever order you like."

In addition to what I covered in the article referenced above, Schmidt had three MVPs and ten Gold Gloves vs. one of each for Brett. Michael Jack led the league in HR and OBP three times and SLG and OPS five times. George led the league in OBP once and AVG, SLG, and OPS three times.

Schmidt's and Brett's peak values were about the same. Brett's counting stats may be better but only because he played a few more years, thanks to the switch to 1B as well as the DH spot the last three years of his career. Schmidt's rate stats (.267/.380/.527) were slightly superior to Brett's (.305/.369/.487).

The fact that I rank Schmidt ahead of Brett in no way minimizes the latter's accomplishments. He is one of the greatest players ever and an inner circle Hall of Famer. Brett was one of my favorite players, and I have a nephew who was named after him. As such, my regard for Brett is very, very high.

Dept. of Irony: Brett and Schmidt were chosen back-to-back in the 1971 amateur draft by the Royals and Phillies. In the SECOND ROUND!

Cyril, I didn't get a chance to read through all your stat analysis, but if you ask Tim McCarver what *caused* it, he'd say it was Gibby's curve ball during that time period. It swooped in like a wiffle ball (looking at first like it might hit you, or be 2 feet off the plate, depending how you batted), and he could place it within a 2-baseball high range on any part of the plate. That would explain higher K rate, lower BB rate and weaker balls in play.

The explanation for Gibson's 1968 success is testament to the validity of "Moneyball".

The key to pitching success, according to Moneyball is 3 primary factors ...

1. Strikout a high percentage of batters faced
2. Walk very few batters
3. Don't give up many HR's

If a pitcher can accomplish the above 3 items ... the only thing that will put him to astronomical statistical heights (as Gibson reached) is to "luck" out and have a low opposing batting average for balls put into play.

I say LUCK because there is really no way a pitcher can control what happens to a batted ball that is in play. What is a base hit up the middle could easily be a routine grounder to short if hit just a little more to the left. A routine flyball to centerfield could just as easily been a double in the gap had it been hit further one way or the other.

Statistics indicate that no pitcher has shown any year-after-year tendency with regards to his success (or lack thereof) for getting batters out on balls put into play. Even Gibson is no except in that regard as the previous year (1967) he actually had a higher batting average than the other pitchers on the Cardinal staff.

The explanation for Gibson's tremendous success in 1968 is the convergence of his skills (the ability to strikeout batters while, at the same time, minimizing walks) and the luck of simply getting batters out when they put the ball in play. When those two things come together in a single season ... you've hit the jackpot.

So why can't pitchers who get "lucky" with balls being put in play get similar numbers? The answer is that not too many pitchers can get the number of strikeouts *AND* the minimal number of walks while, AT THE SAME TIME, getting "lucky."

It's not an accident that most closers are strikeout pitchers. Very little is left to chance when a pitcher can strikeout batters. Once the ball is put into play ... ANYTHING can happen. Pitchers who can be counted upon to strikeout batters out, for the most part, control their own destiny. Pitchers who rely on outs primarily through flyballs and grounders have to cross their fingers.

David Emerling
Memphis, TN