Rockets
No, not pocket rockets, but simply a rocket, any rocket on a board that has an A. I've found this to be true in most lower-limit games (10/20 and below): If there is an A on the board, someone has a pair of aces. If an A is dealt out, it is being played, and so the odds of someone having a pair of aces on a flop that has an A is the same as the odds of one of the remaining three As being among the twenty cards dealt out to the table--in other words, very, very damn likely.
So why, then, does this always seem to happen? I have a pocket pair, say 99. I raise preflop and get a couple of callers. The flop comes A62 rainbow, or AK5 rainbow or something, an A as the only overcard or an A with one other overcard, and no real strong draws to straights of flushes. I bet, knowing if no one has an A, they'll probably fold. I get the same number of callers as I got preflop, so I give up, check-folding just to see at least one but often two people show down aces, often A8o against A4o or some such.
But when I raise preflop with 99 and instead of AK5 the flop comes AK9 and I bet out, everyone folds. Am I that readable? Do I give off "set vibes"? I don't think so; heck, I don't think most people at 10/20 are doing much besides playing their cards. Play any A, and if it hits, check-call to the river. So why is it that whenever I flop a set and there's an A on the board, no one seems to have an A?
I'm at a loss. I understand that the more cards on the flop that hit you, the less chance it is that the flop hit anyone else. But still, an A is an A is an A and whether the 9 flops or not, with an A out there you'd think would hit people just as often as when, intead of a 9, there's a 5 out there.
Oh well. Maybe I've just had an extended unlucky streak in this area. It happens. I'll wait and see. But really, this phenomenon has been going on for a really, really long time.
The Rocket
As people who follow sports have probably heard, Roger Clemens recently signed with the Astros to play another year for $18 million, making him the highest paid pitcher in MLB history.
Roger will turn 43 next season. The level of performance of 43-year-old starting pitchers is pretty consistent--with a few exceptions, those few who pitch at 43 have been consistently bad, even great pitchers (although that's pretty redundant; the only pitchers who will still be pitching at 43 are the great ones).
W L G GS CG SHO IP H HR BB SO ERA *ERA+
3 6 16 16 0 0 87.3 91 7 30 44 3.92 86
0 1 4 1 0 0 9.7 20 5 5 5 16.47 24
8 9 23 23 3 0 150.7 182 9 24 60 3.94 91
6 13 38 25 4 1 173.7 204 23 52 78 5.29 67
0 3 9 3 0 0 21.7 40 5 6 6 9.14 60
7 10 21 20 14 1 163.3 149 0 27 58 2.53 102
(I'm defining age-43 season to mean the player was 43 at some point during the season).
The first guy is Don Sutton, then Steve Carlton, Gaylord Perry, Warren Spahn, Pete Alexander, and Cy Young. All stats are from baseball-reference.com, and *ERA+ for those who don't know is a park-adjusted ERA ratio, which is essentially leagueERA/ERA and then multiplied by 100, so 100 is league-average, higher than 100 is better than average (so a 140 would mean the league average ERA was 40% higher than the player's ERA and a 60 would mean the league-average ERA was 40% lower than the guy's ERA).
These are all hall-of-fame pitchers and as you can see, no one was particularly impressive. In fact, just about all were downright bad. No one was able to pitch more than 200 innings, and just about all were below league average at preventing runs, many way below league average. The only exception was Cy Young, who despite his impressive looking ERA, because it was the "dead-ball" era, was just about league average.
And this list doesn't even include the many great pitchers like Walter Johnson, Christy Mathewson, Lefty Grove, Bob Feller, Bob Gibson, Juan Marichal, Jim Palmer, Tom Seaver, Fergie Jenkins, and Bert Blyleven who all had simililarly poor seasons at an earlier age and never even made it to 43.
So what we have is a whole bunch of guys who were either retired or bad. But who were the effective starting pitchers at 43?
Nolan Ryan was quite good:
13 9 30 30 5 2 204.0 137 18 74 232 3.44 114
as was Phil Niekro:
17 4 35 35 4 2 234.3 225 23 73 177 3.61 104
Charlie Hough wasn't bad:
9 10 31 29 4 1 199.3 167 21 94 107 4.02 99
Tommy John was effective even if he didn't pitch much:
5 3 13 10 1 0 70.7 73 8 15 28 2.93 140
And who can forget about Jack Quinn?
10 11 31 21 8 3 163.7 191 4 36 58 3.41 123
So in all of history, there are basically five starting pitchers who were effective at age 43 (six if you count Cy Young, who was about league-average but had lost a lot of his endurance). Four of those five aren't even comparable to Clemens much at all.
Niekro and Hough were knuckle-ball pitchers, and knuckle-ball pitchers historically have had pretty good longevity, perhaps because it simply isn't that taxing to throw a knuckle-ball, and one doesn't have any velocity to lose--it travels 60MPH anyways.
Quinn, on the other hand, was a spit-baller. The spitball was disallowed in 1920 but Quinn, with a few others, was grandfathered in, allowing him to pitch until a week short of his 50th birthday.
John was a classic junk-baller, striking out very few batters, but also allowing few walks and few home runs. A team has to string together a lot of singles to score many runs without hitting home runs or being given free passes, and John was able to succeed for a very long time on this concept, having one more good season at 44 before sputtering.
The only pitcher who is comparable to Roger Clemens is, of course, Ryan, who is the classic example of a power pitcher. Ryan was nowhere near as good as Clemens, but he was able to be a very effective pitcher well into his forties because his control improved some as he aged and he never lost much off his blazing fastball. Ryan, like John, would have his last truly effective season at age 44.
So what does this mean for Roger and the Astros? Was it a dumb signing?
Although historically there have been a fair number of pitchers who have fallen off a cliff, so to speak (Gibson and Carlton come to mind), barring injuries, pitchers typically regress a little bit each year until they finally slip below a certain level, at which point they are either released or they retire.
As evidenced by last season, Clemens is still performing at a very high level. He also has become less reliant on his fastball (which has lost a few MPH), and is depending more on his splitter. This is a good thing, because what causes many pitchers to become ineffective is they lose their top-flight fastball and have nothing to replace it with.
There are various projection tools out there, and one, ZiPS, projects Roger to start 32 games, going 14-9 with a 3.75 ERA and 196 Ks in 204 IP (the 3.75 ERA would translate into a 115 *ERA+ assuming last year's leagueERA). That's not a bad line, and were he to be approximately that effective, he would rate comparably with pitchers who are earning in the $8 - 12 million range. Obviously, were he to repeat his 2004 performance, he'd be in the $13 - $16 million range with Pedro Martinez, Curt Schilling and Randy Johnson.
But. He's getting paid $18 million. And no one thinks he'll improve in 2005. So the Astros overpaid, right? Not necessarily.
If they were paying for just performance, there's no doubt they overpaid, probably by about $6 million. But Clemens brings a lot more than just performance to the 'Stros. Last year when he signed, season-ticket sales shot through the roof. Attendance for the entire year was up about 26%. Yes, they made the playoffs, but they've been competetive every season and still never matched 2004's attendance, a year for which during two entire months, July and August, people legitimately believed the Astros were out of any postseason races.
The attendance had been on a clear and consistent downward trend ever since the spike when the new stadium was completed for the 2000 season, and the reason that trend turned around was Roger Clemens. In 2004, Houston had become a baseball town. So Roger is a major gate attraction, increasing team popularity, ticket sales (and therefore concession sales), team apparel sales and the overal marketability of the franchise.
This past offseason also saw the Astros lose Carlos Beltran and Jeff Kent, two main contributors to the 2004 season, and so going into the 2005 season, were the Astros to also lose Clemens, who grew up in Houston, not only would they be a non-contender, the fans would have no reason to come out to the ballpark much less follow the team. It'll still be tough for them to contend, what with Biggio and Bagwell another year older and Berkman out for at least the first month, but at least they have a gate attraction.
So for many teams, perhaps paying Roger Clemens $18 million would be overpaying. But for Houston, Roger is worth every penny.
damn fine baseball post - i concur with your analysis 125%. kudos.
Posted by: iggy | January 26, 2005 at 18:09