Happy Thanksgiving! Regularly Scheduled Articles Will Resume Monday, December 1
July 16, 2003
Lies, Damned Lies
PECOTA Mid-Season Review
Watch SportsCenter this time of year, or read the Sunday baseball page--that's the one with the long list of players sorted by their batting averages--and you're sure to see plenty of stories about what a wonderful, surprising baseball season this has been. Why, who would have thought that Dontrelle Willis would have been drawing Mark Fidrych comparisons, that the Royals would be 10 games over sea level at the Break, that Melvin Mora would be an MVP candidate, that Esteban Loaiza would be the best arm in the American League? Perhaps there's some Joe Namath among you, some Nostradamus, some Miss Cleo, but we certainly didn't.
Then again, how would you have felt if you came across headlines like these:
With 30 teams, and 750 ballplayers divided among them, every baseball season is sure to have its share of surprises. But all things considered, this has been a remarkably predictable baseball season. Compare the All-Star break standings to those predicted by PECOTA before the season--you can find the relevant links here and here--and you'll get a correlation on winning percentage of .74. Pretty good, huh? PECOTA must really be onto something.
Well, it seems pretty good, until you consider that the correlation between the year-end 2002 standings and the year-to-date 2003 standings is even greater--around .79. Make no mistake: that's an unusually high figure, as we can see by comparing it to the correlations from previous seasons.
Year-over-year correlations in Winning % 2003 (partial year) v 2002 .79 2002 v 2001 .69 2001 v 2000 .50 2000 v 1999 .47 1999 v 1998 .58 1998 v 1997 .34 1997 v 1996 .35
The last time that the correlation was so high was between 1965 and 1966--that was in a league with a balanced schedule, 20 teams, and no hint of free agency. Sure, there's still plenty of baseball left to be played, but if anything, you'd expect partial-season results to be more erratic than year-end results.
So what does it all mean? Well, I think there are about three lines of thought here:
Although I think there's an element of truth in each theory, you can pretty much tell from the phrasing which one I'm most inclined to believe. Whether it's a sign of improved business models or not, we haven't seen any teams engaged in a full-scale bailout this year, as the Marlins performed after 1997 or the Padres after 1998. With the possible exception of the Phillies, we also haven't seen any teams kick it into the turbocharged, win-at-all-costs gear that Tom Hicks made famous a couple of winters ago. Most teams stand at about the same place in terms of their willingness to spend as they did a year ago, and most teams' time horizons have not shifted dramatically.
On the other hand, there may not be as many possibilities for arbitrage as their once were. Between 1997 and 1999, the A's advanced from 65 wins to 87, simply by giving the guys in their farm system a chance, and picking up veteran spare parts that nobody else wanted. A team like the Tigers or the Reds might have their hearts in the right place, but there are more teams competing for "unappreciated" talent, and fewer teams willing to take on a Bobby Higginson- or a Barry Larkin-type of contract. The hill has become that much steeper to climb.
There's still plenty of baseball left to be played, of course, so all of this discussion may be premature. In the meantime, let's take a division-by-division comparison of the PECOTA projected standings against the year-to-date actuals. Actual figures are prorated to 162 games, and we'll look at runs scored and runs allowed projections as well as the win totals.
AL EAST Actual Projected +/- W RS RA W RS RA W RS RA NYA 99 880 719 109 986 688 -10 -106 +31 BOS 96 998 826 104 914 673 -8 +84 +153 TOR 84 948 878 80 838 846 +4 +110 +32 BAL 73 819 824 69 674 788 +4 +145 +36 TBA 56 681 893 57 674 923 -1 +7 -30
We boldly predicted that the AL East standings would resolve themselves the same way that they have for each of the past six years, and though the Jays seemed poised to throw a wrench--or is that a B.J. Birdy?--in those plans for a while, it's a bet that seems even safer now than it did at the start of the season. Sure, the Red Sox still have a shot, but with the Yankees' medical outlook improving, and George Steinbrenner's wallet yet to make its presence felt, New York is in the driver's seat.
Don't think they'll be winning 109 games, though.
PECOTA's Greatest Hits
PECOTA's Greatest Misses
AL CENTRAL Actual Projected +/- W RS RA W RS RA W RS RA KCA 90 840 840 64 762 951 +26 +78 -111 CHA 78 677 701 86 850 797 -8 -173 -96 MIN 77 749 812 84 799 768 -7 -50 +44 CLE 71 695 784 67 732 879 +4 -37 -95 DET 44 523 812 55 647 913 -11 -124 -101
They haven't played up to their April numbers, and they've given up as many runs as they've allowed, but the Royals remain the biggest surprise in baseball, and there's no doubt that this is a more talented team than we expected it to be.
Still, Royals fans have some cause for concern. Like last year's Angels, the Royals have been superbly good with runners in scoring position:
Royals, overall: .272/.338/.429 Royals, RISP: .301/.378/.455
As the Angels proved, sometimes flukes like that can maintain themselves over a full season, but more likely than not, the Royals are going to give something back. As a result of their performance in clutch situations, the Royals have scored 35 more runs this season than their raw numbers project them to. Combine that with their favorable record in one-run games, and this is a team that's been lucky twice over.
On the other hand, with a seven-game cushion, and the White Sox and Twins apparently headed in reverse, that might be enough. It's better to be lucky than bad.
AL WEST Actual Projected +/- W RS RA W RS RA W RS RA SEA 101 817 606 90 788 705 +11 +29 -99 OAK 94 746 645 89 788 710 +5 -42 -65 ANA 86 792 681 82 781 772 +4 +11 -91 TEX 66 805 1007 83 858 833 -17 -53 +174
Here's what I wrote about the Mariners in March:
A lot of people are ready to write the Mariners off; of the 30 or so people who filled out prediction forms at our Chicago Pizza Feed last weekend, just a couple picked the Mariners to win the division, while a couple more expected them to sneak in as the Wild Card. Many of their vulnerabilities were exposed last season: Freddy Garcia (190, 3.81), worn down by heavy usage, pitched horribly in the second half, Edgar Martinez (.274/.394/.483) suffered through an injury-riddled campaign that looks like the beginning of the end, and Jeff Cirillo (.255/.306/.348) did everything he could to counteract Andres Galarraga's example that road stats for Rockies shouldn't count...But the Mariners are still a good team...Because the starting lineup is old, there aren't many breakout candidates among the bunch, but if the A's and Angels both fall back to earth, slow and steady could win the race.
Lest I make too much of this, I should remind everyone that I didn't maintain my faith in the Mariners in our more subjective projection exercise, bucking the PECOTA projection and picking the A's to hold on to the division in our pre-season projection roundup. But the Mariners are a strong, healthy, balanced team, well ahead in their division without any individual performances that are obviously flukes (well, excepting Shigetoshi Hasegawa). I should know better than to count out any team with Billy Beane at the helm, but the Mariners' hold on the AL West seems solid.
NL EAST Actual Projected +/- W RS RA W RS RA W RS RA ATL 106 881 746 85 763 717 +21 +118 +29 PHI 92 771 615 94 825 672 -2 -54 -57 MON 84 736 717 82 760 756 +2 -24 -39 FLO 84 771 738 79 710 731 +5 +61 +7 NYN 70 678 820 82 691 680 -12 -13 +140
There's been a lot of talk about the Braves, but the Phillies might be the more interesting team. Their season has been thought of as a disappointment because the Braves pulled so far ahead so early, but the Phils are on pace to win 92 games. Compare and contrast their situation with that of the Cubs, a nominal playoff contender because of their weak division, who have much of Chicago clamoring for a future-is-now deadline deal even as they sink back to .500.
The Phillies have far and away the best pitching in the division, perhaps the best pitching in the league. If Pat Burrell can get it turned around, this is a very dangerous team.
NL CENTRAL Actual Projected +/- W RS RA W RS RA W RS RA HOU 86 815 696 89 830 731 -3 -15 -35 SLN 84 920 815 86 789 725 -2 +131 +90 CHN 81 719 722 83 698 672 -2 +21 +50 CIN 75 756 951 79 734 762 -4 +22 +189 PIT 73 721 794 78 734 769 -5 -13 +25 MIL 64 744 883 68 639 791 -4 +105 +92
In terms of wins and losses, PECOTA has done remarkably well with the NL Central: Each team is on pace to finish within a few games of its wins projection (albeit in every case on its downside). The runs scored and runs allowed numbers, however, tell a much different story. This is a three-tiered division: the Astros and Cardinals are serious contenders, the Cubs are waiting until next year, while the Reds, Pirates and Brewers mostly serve to provide a bottom as generous as J-Lo's.
NL WEST Actual Projected +/- W RS RA W RS RA W RS RA SFN 98 769 689 84 754 716 +14 +15 -27 ARI 90 779 705 86 699 649 +4 +80 +56 LAN 85 571 531 82 674 669 +3 -103 -138 COL 84 894 878 82 935 930 +2 -41 -52 SDN 59 687 856 72 656 764 -13 +31 +92
This might end up being the most interesting division in baseball. Though the Giants have lead from the get-go, the top four teams are separated by just a couple of games in the Pythagorean standings. What's more, these are all tremendously different teams. The Giants are as dependent as ever on their superstar; the Diamondbacks have made it here largely with performances from unknowns and underappreciateds. The average Rockie game involves about 11 runs scored; the average Dodger game, fewer than seven--sure, you've got park effects involved, but that's still a lot of fun. Now that Rickey Henderson has joined the party, East Coasters may want to reschedule any 8 a.m. conference calls they have lined up for late September.
If we combine the season-to-date standings with the PECOTA projections for the balance of the season, here's what we come up with for a revised list of playoff teams and their win totals:
AL: Yankees 103, Red Sox (wc) 99, Mariners 96, White Sox 81
I'll buy most of that, including the possibility that the AL Central is won by a team without a winning record, though I expect that the Cardinals will pick up a pitcher or two at the deadline and win the division handily. Enjoy the rest of the baseball season, and we'll check back in on PECOTA's progress in October.