User Comments, Suggestions, or Complaints | Privacy Policy | Terms of Service | Advertising
Buy MLB playoff tickets, plus 2011 World Series, 2011 ALCS tickets and NLCS game tickets. We also have Texas Rangers playoff schedule, tickets to Red Sox games and Yankees game tickets. Plus, buy Phillies baseball tickets, Tigers playoff tickets and the biggies like ALDS baseball tickets and 2011 NLDS tickets. |
Demarini, Easton and TPX Baseball Bats
|
AllianceTickets.com has cheap MLB Tickets. Get all your Colorado Rockies Tickets, Seattle Mariners Tickets, San Francisco Giants Tickets and all your favorite baseball tickets here. We also carry cheap Denver Broncos Tickets, Seattle Seahawks Tickets and Denver Nuggets Tickets. |
Page rendered in 0.5479 seconds
54 querie(s) executed
Reader Comments and Retorts
Go to end of page
Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.
1. bunyon Posted: November 10, 2004 at 03:02 PM (#960449)Thanks, Klesko, you oaf.
This is kind of interesting. How close did Don Sutton ever come? He doesn't seem like the kind of overpowering guy you think of throwing a no-hitter, but then underpowering guys to it too.
Sid Fernandez 6.85 300 .839 0
According to James, El Sid is 5th on the list of "Most likely to throw a no-no but didn't," behind Sutton, Martinez, Clemens, and McDowell.
But you also have to be able to throw a complete game to throw a no-hitter, and that HAS to knock Fernandez well down on the list, doesn't it? Fernandez threw a grand total of 25 CGs in his entire career, less than 1/10th of the total tossed by Steve Carlton (254) in his career. Yet by James's method, Carlton ranks just below Fernandez in likelihood to throw a no-hitter.
I guess that's a quibble. It's an interesting topic and a neat little take on it. And in general, the top of the list is full of guys who you'd probably name as the most likely to throw some classics.
Well, I'm not doing anything productive this week anyway . . .
Depending on how you count it (& ignoring for the moment the fact that retrosheet doesn't have box scores for the NL for his first 3 years), he tossed 3-4 one-hitters. The closest he came in those was on 8-18-77 when Marc Freakin' Hill hit a two-out single off of him in the 8th.
Came 5 outs away in a 5-1-69 game against the Giants before Jim Davenport doubled off of him. Jim Wynn got the sole hit against him on 6-19-71 in the 6th inning. He debatable one-hitter is when he went 10 innnings in a 13 inning game against the Expos, with Bob Bailey get the only hit off of him (in 7th inning). Montreal went on to get two more hits & the game's only run though.
Maybe, but maybe not.
If El Sid is tossing a no-hitter through 6 or 7 innings, how many managers will take him out of the game?
I would further guess that if you were to try to adjust for liklihood of throwing a CG, you'd want to weight for average IP/start. If someone has a no hitter after 7, they'll stay in.
I'll add to my previous post that, yes, a pitcher may stay in for an 8th or 9th inning if he has a no hitter working. However, if doesn't have the stamina to work those innings, is he likely to shut down the opposition. IOW, I don't think you can simply say that one guy didn't complete many games so he'd never throw a no-hitter. But you can say that if a guy didn't throw a lot of CG, he wasn't likely to stay effective through nine and, thus, you'd have to lower his expected no-hitters.
Which, I think, was Sam's original point.
Plenty, because he'll have walked five through 7 IP, and then put the first two on with a HBP and another walk in the 8th. He'll also be on the verge of collapsing, and need oxygen on the bench after being removed. ;-)
Ah, Sid -- I kid you because I love you. Looking back, your misadventures in the 6th and 7th innings were the good old days!
5-15-87, tossed 5-no hit innings vs. San Fransisco, K'd 7, walked 4. Huh. Wasn't even lifted for a pinch hitter or anything. His last inning went K-W-FO8-FO8.
8-31-88 tossed 1 inning, K'd 1, no walks. Well that hardly counts. Must've been injured or something.
That's it.
He had a no-hitter (perfect game?) through 9 in 1995, gave up a hit in the 10th.
Is there any sort of adjustment for predicted no-hitters versus actual no-hitters. Neyer touches on using 26 outs rather than 27. But I am wondering about how many no-hitters there have been in baseball as a whole, versus the cumulative expectations of the players this way. I'm sure someone has looked at this already, but has there been a reasonable number of no-hitters thrown to what can be projected this way?
Ryan 2.715
R. Johnson 1.213
W. Johnson 1.166
Seaver 1.127
Clemens 0.936
Martinez 0.902
Sutton 0.891
Maddux is at .580, passing Whitey Ford. I didn't check Smoltz.
Wouldn't it depend on the pitch count? I don't have any supporting evidence, but I have to think that the number of pitches thrown during a 9-inning no-hitter would be roughly the same as a 7-inning quality start
However, if you take out his first two seasons (Rule V year and year in which he only pitched in 15 games), then his number jumps up to 0.32% chance. That is, if he starts 30 games a season for the next 9 seasons, he should throw a least one no-hitter.
I wanted to tweak this equation to apply it to perfect games, so I changed it to:
(3 * IP) / ((3 * IP) + H + BB + HBP)
(If I've made any huge logical errors, please let me know.) I don't have the data for reached on errors, so I won't include it and hope it works right anyway.
I wanted to know, "How likely is it for a pitcher to throw 12 perfect innings?" For this I'll use a hypothetical pitcher; we'll call him Harvey. Harvey's numbers are:
2235 IP, 2798 baserunners allowed
So, 3 * 2235 / (3 * 2235 + 2798) = 0.70556663
I tried to raise this number to the 36th power, and my calculator said the answer was this:
3.526711266e-6
Did I do the calculation right? And if I did, can someone tell me what the hell this number means? I'm going to go out on a limb and say that it means that 36-up, 36-down doesn't happen often.
I think the main reason Sutton is a sleeper is not the starts. It's that we tend to forget (at least for the purposes of this kind of thing) he pitched in some extreme pitcher's parks.
Authorities believe doughnuts were involved.
Thanks. So Harvey's chances of 12 perfect innings would be around 285/283,550 or 1 in 995.
I went to that game, after making the same prediction.
This is the nature of extremal statistics: they are usually dominated by fluctuations, not averages.
Perhaps the best way to account for that would be to do a season-by-season prediction. Of course, the whole point of the exercise was to come up with a quick answer, but it could be improved relatively easily.
Sudden Sam threw 2 consecutive 1-hitters in 1966
(which is apropos of nothing except I was at the 2nd one)
David Cone almost did the same thing in 1991. Shortly after Tim McCarver predicted, on the air, that Cone was throwing so well that he would pitch the Mets' first no-hitter before the season was out, Cone produced these two gems, back-to-back:
Sept. 14: 7 1 2 2 4 7
Sept. 20: 9 1 0 0 1 11
Then in his last start that year (Oct. 6) he did this against the Phils:
9 3 0 0 1 19
Yowsa.
That would make VIrgil Barnes his cousin as well. Gammons once mentioned that he was related to Ethan Allen. Allen being the player and longtime college coach, not the Green Mountain boy.
In fact, since every pitcher who throws for more than one season has a "peak" that exceeds his career average by some degree, looking season-by-season has the effect of raising EVERYONE's no-hit percentages.
I ran numbers just for Nolan Ryan and Sandy Koufax. Ryan goes from 2.71 expected no-nos to 3.15. Koufax enters joins the four above him in the "expected to throw one" category, going from 0.92 to 1.05.
Of course, if you break it down into game-by-game predictions, everyone will be spot-on.
The year in which Nolan Ryan was most likely to throw a no-hitter -- 1972 (0.38 expected), he didn't. He did, however, throw one each in his second most likely year (1991, 0.25) and third most likely (1974, 0.23). The year he threw 2 no-nos, was his 10th most likely (1973, 0.14). 1975 (0.07, ranked 18th in his career) was the year with the lowest probability in which he threw a no-hitter.
I ran the numbers on two pitchers with the exact same career hit rate, and a similar number of starts: David Cone and Whitey Ford. Neyer has them very close with his stat (Cone has .582, Ford has .575).
The year-by-year numbers i got were:
Cone: .703
Ford: .637
The much larger difference comes from the fact that Ford was extremely consistent throughout his career (H/9 variance of .59), while Cone was extremely inconsistent (H/9 variance of 2.20).
I think the latter is cool too.
One other quibble to add to that - what happens if a Ryan or similar pitcher really has his stuff that day? In "The Umpire Strikes Back" Luciano mentions that he ump'd one of Ryan's no-hitters and said hitters were thanking him when he called them out on strikes. Had decent control that day and his curve was really working. I guy being "in the zone" could increase his odds, which may explain why a Ryan or Koufax has far more no-hitters than this formula projects for them.
Sudden Sam & Cone got nothing on Dave Stieb. Back-to-back games where DS allowed his only hit with two outs in the ninth. One of those came on the flukiest hop that ever fluked.
And yes I know, Johnny VanderMeer owns them all.
On the topic of Sid Fernandez leaving two games with no-hitters: The same problem -- a troublesome left knee -- forced him out of both games. In the '87 game against the Giants, his knee buckled as he started to throw the first pitch of the sixth inning to Mark Wasinger. In the '88 game, he left the game after one inning with a sore left knee, prompting rumblings about his unhealthy weight.
Right. Actually, the point I was going to make but somehow forgot to was this:
Thirty expected no-hitters, 37 actual, with Nolan Ryan creating most of the discrepancy.
One would expect that if you went year-by-year for all of them, you'd get a whole lot closer to 37.
I originally thought those were back-to-back starts but checking retrosheet they weren't. In between those two gems, he gave up 7 runs in 4.1 IP against Houston.
This was right when Randy Johnson was going through the most God-forsaken stretch of offensive "support" of any pitcher you'll ever find. Four straight times the D-backs were shutout when he pitched.
That's bad, but it's only tied for the worst that I know of (happened to Derringer in '33, the Giant who lost 18 in '68 & a Senator in the early '60s, and I'm sure to numerous others).
What sets Johnson apart is: 1) the environment in which he got screed - far higher scoring than the '68 NL, or even the '33 NL. 2) Not only were the D-backs shut out, but they were no-hit, one-hit, two-hit, and three-hit (neat how it worked out in that order, huh?) 3) He pitched so wildly lights out -
32 IP, 19 Hits allowed, 10 walks, 54 K's, 6 runs allowed, 5 earned, for an ERA of 1.41.
Zero wins, four losses. Cost him 20 wins on the year. God, it's bad enough you had to give him that face, did you have to do that to him, too? He must've defiled a whole bunch of temples in a previous lifetime.
Added bonus: not only were the no & two hitters by Jimenez, but it was Ron Frickin' Villone with the one-hitter. Ouch.
I am not adding to the conversation other to say, 'I agree.'
"which may explain why a Ryan or Koufax has far more no-hitters than this formula projects for them. "
I think your point must be considered.
"One would expect that if you went year-by-year for all of them, you'd get a whole lot closer to 37."
Maybe Koufax's discrepancy more so than Ryan, because he wasn't as dominant in his early years. Both of them probably had a high difference in Home and Away Opposing batter's average as well as Day/Night splits. (How about a twilight game against them?)
"Of course, if you break it down into game-by-game predictions, everyone will be spot-on. "
You got us there. I guess the question becomes under the right conditions, who are the pitchers most likely to pitch a no-hitter? And who didn't?
slightly below the heartbreak of psoriasis
Yet the Orioles got skunked. Go figure.
What's the best hitting team to ever get no-hit & who's the pitcher that did it. Or to make it even more precise, what's the best hitting line up that ever go no-hit?
His game scores for that four-game losing streak: 85, 77, 78, and 72. Jesus Ethelred Christ that's good.
I'll give an opening bid of the July 13, 1991 Oakland A's in the 4-pitcher no-hitter by the Orioles. The Top 5 in that batting order were Willie Wilson, Dave Henderson, Jose Canseco, Harold Baines, and Mark McGwire. Terry Steinbach and Rickey Henderson pinch hit.
I went ahead and made the year-by-year list. The rankings were different, so it's not the exact same group of players, but my top 40 pitchers threw 35 no-no's, vs. a predicted 36.594. I think a refining of the formula (maybe a slightly lower exponent) would improve my method even more.
Interestingly enough, the year-by-year method actually changed the answer to the orginial question. Sam McDowell edged out Sutton for the honor of "most likely pitcher to throw a no-hitter who never threw one". Very suprising.
More, importantly, Jesse Barnes is Bill James's cousin?
James has a short essay in one of his books about how he found out that he was related to the Barnes brothers after the publication of the original Historical Baseball Abstract. He mentions that they were from the same county as him in that book, but had no idea that they were related until later.
If I have a moment, I'll see if I can find the article.
Hideo Nomo throwing the no-hitter in Coors field has to rank up there. Any lineup in Coors is a pretty great lineup.
The Yankees from last year were pretty good. Soriano, Jeter, Giambi, Posada, Ventura, Mondesi, Matsui.
That was quite a 3 game series. In the second game, the Giants hit 5 HR in one inning. In the third game, Mays hit 4 HR.
This isn't true.
The number of predicted no hitters would be higher than the number of actual no hitters.
Using the formula {(3*IP)/[(3*IP) + H]}^26 for each game an actual no-hitter is thrown you get:
1^26 = 1
In other words, each actual no hitter would give you a theoretical no hitter. The problem is that other games also give you a chance at a no hitter. For instance, an actual 1-hitter:
(27/28)^26 = .39
When a pitcher is pitching at "1-hitter quality" he has a 39% chance of throwing a no hitter in any given start.
Using the results of each game as an indicator of a pitcher's ability in that game overestimates the fluctuation in ability of a pitcher (just like using a career rate underestimates the fluctuation) and thus overestimates the number of predicted no hitters.
If you find the period of time (maybe a season) where a pitcher's results acurately predict the number of no-hitters a pitcher will throw, you might also have the unit of time where fluctuation in a pitcher's results most accurately reflects the amount of fluctuation in real ability.
Does that make any sense?
#44: I was watching on TV. Yes, it was Denver, and it was pretty amazing - but it was a cold, wet night in Denver. I don't think the park factor was up to its usual level that night.
One, as has already been mentioned, except under rare circumstances, a pitcher pitching a no-no stays in the game. Two, even if a particular group of pitchers tend to get tired after X number of pitches and therefore would be less likely to sustain that no-no in the late innings, if you ARE pitching a no-no, again, except under rare circumstances (lots of errors, walks, HBP's, or deep counts), your pitch count is likely to be quite low.
As far as using peak years, etc., to re-do the calcs, that is probably not going to work as the only thing that matters, in terms of a pitcher's chance of throwing a no-no, is his "true hit rate." That probably doesn't change much over a pitcher's career, other than according to a typical aging pattern. The year-to-year hit rate fluctuations you see in a pitcher's career are mostly random fluctuation. Those random fluctuations do not change a pitcher's chance of throwing a no-no.
The fact that there have been 58 expected and 57 actual no-no's among the pitcher's isolated in the article suggests (although there is obviously a healthy "confidence interval" surroundign that 58) that pitcher's DO NOT have good days and bad days other than through random fluctuation around their true talent level which is probably pretty stable across starts.
Finally, there is no evidence (in fact, there is evidence that this is NOT true) that pitchers have certain types of batters or teams that are more or less "optimal" for them (or even for their general type or style). That is probably another of scores of "myths" in baseball. Again, the 58/57 numbers suggest (albeit weakly) that that is not the case...
even if pitchers don't have certain "types" of batters that are more/less likely to get a hit against them, they certainly DO have certain batters that are more likely to get a hit off of them. Wade Boggs is more likely to get a hit against them than Kevin Elster. Since, a .270 lineup has soom .250 hitters and some .290 hitters, a pitcher's chance of throwing a no hitter is somewhat less than it would be against a team of all .270 hitters:
All .270's: .73^26 = 2.80x10^-4
half .250's, half .290's = (.75^13)*(.71^13) = 2.77x10^-4
The difference is obviously small. No hitters are 1% less likely with the varied lineup.
If you use .230/.310 you get 4% fewer no hitters.
The big change is when you use a different # of outs:
# of outs, not hitters per 10,000 games (.270 baa)
26.00, 2.80
26.25, 2.58
26.50, 2.39
26.75, 2.20
27.00, 2.04
We should check Rob's assumption of 26 outs. How many DP's, CS are there likely to be in a no-hitter?
I agree with most or all of that -- except it isn't true with Sid Fernandez. "Quite low" pitch counts and Sid Fernandez did not go together. He was a deep count machine, whether he was pitching effectively or getting ripped. That usually included a lot of walks, too.
Maybe what I really meant to point out was that the method simply overstates El Sid's chances of ever throwing a no-hitter -- i.e., that he's the outlier in an otherwise reasonable assessment. Here are the "leaders" from that list in fewest career CGs:
1. Sid Fernandez 25
2. Pedro Martinez 42
3. John Smoltz 47
4. David Cone 56
5. Bob Turley 78
6. J.R. Richard 76
7. Andy Messersmith 98
8. Randy Johnson 92
As you'd expect, it's the modern guys who dominate the list. But El Sid is the one who really stands out as a guy who just couldn't be expected to go nine, even if he was carrying a no-hitter into the late innings.
El Sid averaged 6.2 IP/start and 25.6 batters/start over his career. I think it's safe to assume that in some of those starts he was pulled early and that, on average, when he was pitching well he faced more than 25.6 batters in a game. He walked 3.4 batters/9 IP so in order to get 26 outs in a no-hitter he'd need to face 29.4 batters. I think he could manage that. He averaged over 27 batters per game in his best seasons.
IP: 2414.3
H: 2092
H/9: 7.80
GS: 361
No-Hitter probability: .493
So somewhat lower than .514 reported, at least enough to knock him off that list.
Of course, this may not be fair, as pitchers in earlier eras often pitched both as starters and as relievers when needed. And going through and allocating actual hits and innings to starts and relief appearances would be troublesome. But where it's easier to do (as in Smoltz's case), there seems no reason not to do it.
That address again:
Chris Rainey
Box 37
Clifton, Ohio 45316
Hurry, order before midnight tonight!
Well, of course he could do it. The issue here is one of relative probabilities. Compared to the other guys on this list, El Sid was considerably less likely to do it than his pure numbers of hits and innings pitched would suggest, because he is less likely to get the complete game that's a prerequisite for a (solo) no-hitter.
My his method, I get a .0037 expected perfect games for Bumpus. That's second (or possibly third) place, behind Iron Davis.
Unless I misunderstood James methodology.
Bumpus: 9.19 H/9, 7 GS, .0038 expected no-hitters
Holloman: 9.51 H/9, 10 GS, .0039 expecte no-hitters.
Iron Davis: 9.19 H/9, 22 GS, .0108 expected no-hitters
So, Bumpus apparently retains the least-likely no-hitter crown (barely), as well as the fewest career starts with a no-hitter crown (since Mike Warren really has 27, not 7, starts and Bud Smith has 24, not 4), but loses the highest H/9 with a no-hitter crown, to I-don't-know-whom.
(1) He's one of two pitchers to pitch a no-hitter in his first start.
(2) He's the only pitcher to do it in his first game.
(3) He was the first pitcher to do it on the last day of the season.
(4) He did it on the latest date of anyone, including Don Larsen.
(5) He obviously gave up the fewest hits per game of any starting pitcher in any season, and had the lowest opponents BA.
(6) He is the only person to pitch a no-hitter in his only start of the season.
and
(7) He was the last ML pitcher to throw a no-hitter at the 50-foot pitching distance. They changed to 60ft 6in the next season and goodbye career.
He also holds a lot of freak records beside you.
I also have a bunch of Freak Records beside me...usually they are filed next to my Frankhauser LP's.
You must be Registered and Logged In to post comments.
<< Back to main