September Stats, Winter league myths

SEPTEMBER STATS AND WINTER LEAGUE MYTHS

Every year we read with amusement those scribes who come out with lengthy articles extolling this player or that one based on his performance in September or in the winter leagues. Lately we have even seen subscriptions to "Winter League Reports" and other scouting reports.

The main problem in relying on these time periods is that they are so short. In September, most players will get significantly less than 100 at-bats. If a player gets 80 at bats in September, the difference between 21 and 24 hits is the difference between .263 and .300. As Crash Davis says in Bull Durham "Just one dying quail a week." Also, depending on where a team is in the standings a player may see a lot of AAA pitchers in September.

Thus, if someone publishes a book and devotes space to September stats you can pretty well ignore it. Can they find examples of players who do well in September and carry it over to the next year? Of course. The real question is whether you can do this with any consistency. Let's take a quick look at some September 1995 performers.

Juan Guzman rebounded from an ERA of 6.32 to win the ERA title in 1996. In September 1995 he was 1-4 5.31 and a WHIP of 1.41. Better than the season but nothing to indicate 1996's performance.

Frank Thomas hit .242 in September 1995. Of course there were the usual predictions of his demise.

Troy O'Leary hit .197 in September 1995. Obviously indicating a player to avoid in 1996.

NL MVP contender Ellis Burks hit.279 with 1 homer in 68 at-bats in September 1995. Obviously his power was gone, or so the pundits said.

Vinnie Castilla's 32 homer 1995 season was deemed a fluke after he hit 2 homers in 107 at-bats last September. Some fluke.

How about Todd Hundley in September 1995? Try .271-2-8 in 70 at-bats

Andres Galarraga hit for his lowest full month average and fewest homers in September 1995. He was obviously not going to win the NL home run crown in 1996.

By now you are probably laughing at some of these examples, as you should. Of course there are other players whose 1996 season more closely mirrored their September 1995 stats. The point is that, statistically speaking, there is no basis for relying on September. In fact, for 1996 it was just the opposite.

Let's cut to the chase and do something extremely radical. Let us look at the entire league population for September/October 1995. In 1996, the offense in the AL exploded. In 1995 the AL ERA was 4.71. For 1996 the AL league ERA was 4.99. The AL league batting average went from .270 in 1995 to .277 in 1996. Thus, in September 1995 we should have seen better than average offensive output from the league as a whole if September is a reliable indicator of future perfomance.

You probably already know the answer. (Note: we are not counting strike shortened April 1995 in our monthly comparisons). In September/October 1995 the AL hit.263, it's lowest monthly average for the year. The league hit the fewest homers of any month and had the lowest ERA (4.38) of any month.

How about the NL league? You have seen all the pundits discuss the great offensive explosion in 1996. Would it surprise you to learn that in NL ERA was 4.18 in 1995 and 4.21 in 1996. The league batting average actually dropped from .263 to .262! Yet in September 1995, the NL had it's highest monthly batting average (.267) and second highest ERA 4.30). In other words, there was no offensive explosion in the NL but the September stats would lead you to believe that there would be.

Now, one more time to drive the point home. September is not only no better than the whole season, because it is such a short period, it actually distorts the picture of what the majority of players will do the following year. Will there be examples of players who follow form? Yes! However, they are in the minority. Remember, when you are dealing with a universe of 700 players you can be right on 200 and that is nothing to brag about.

It means you were off on 500!

Winter leagues generally have the caliber of good AA or decent AAA leagues. Thus, why should 100 at- bats negate the performance a player has put up at those levels the last 3 years? The fact is that it should not.

For another point of reference, look at the 1993-94 Winter All-Star team as selected by Baseball America:

C Carlos Delgado
1b Alan Zinter
2b Luis Sojo
3b Russ Davis
SS Orlando Miller
OF Sil Campusano
OF Juan Gonzalez
OF Ted Wood
DH Nelson Simmons

SP Apolinar Garcia
SP Vincente Palacios
SP Carlos Pulido
SP Ed Vosberg
RP John Hudek

Delgado made the Blue Jays in the spring and then fell on his face. Zinter hit poorly at AAA. Sojo played decently for Seattle. Russ Davis spent another year at AAA. Orlando Miller did not hit well in AAA despite being in the PCL. Campusano, Wood and Simmons did not make it to the majors. Gonzalez had a let down from 1993. The Indians moved Garcia to long relief at AA. Palacios was 3-8 3.44 for the Cards while Pulido went 3-7 5.98 in Minnesota. Lefty Ed Vosberg couldn’t crack the Oakland staff with it’s stellar 4.80 ERA. Hudek had a good first half but faded.

Now, let's fast forward to the winter of 2002-03. Here are the winter all-stars as determined by Baseball America followed by their 2003 major league stats.

C Eliezer Alfonzo (DNP)
1b Ken Harvey (.266-13-64-2)
2b D’Angelo Jimenez (.273 14-57-11)
3b Jose Castillo (DNP)
SS Jose Reyes (.3907-5-32-13)

OF Luke Allen (.000-0-0-0)
OF Ruben Mateo (.242 3-18-0)

OF Rene Reyes (.259-2-7-2)
DH Julio Zuleta (DNP)

SP Edgar Gonzalez (2-1 4.91 ERA)
SP Kiko Calero (1-1 2.82 ERA)
SP Bobby Jenks (DNP)

SP Oscar Villarreal (10-7 2.57 ERA)

RP Arnie Munoz (DNP)

OK, a little better than past years but still nothing to go crazy over. Harvey was a known commodity and Jose Reyes was the next great shortstop by that point. D’Angelo Jimenez had a fine year in 2003 but the biggest change was going from a career back-up to a starting role. Only his 2003 HRs were really a surprise. Oscar Villarreal parlayed a strong winter and strong spring into a long relief role with Arizona that resulted in a lot of wins. Luke Allen and Ruben Mateo still failed to establish themselves as major league players despite a great winter.

Here are the AFL hitting and pitching leaders the last few years:

November 2001:
Derek Nicholson .438-3-18-0
Orlando Hudson .426-2-20-8
Bill Ortega .387-2-20-4 (note Carl Crawford did finish 4^th – we will give you that one)

Kyle Kane 0-0 3 SV 1.80 ERA
Mike Gonzalez 0-1 1.99 ERA
Eric Cyr 2-1 2.180 ERA

November 2002:
Ken Harvey .479-7-34-0 (WOW)
Tim Olson .374-1-18-6
Adam Morrissey .371-4-14-4

Dave Sanders 6-0 0.75 ERA
Josh Stewart 1-0 0.81 ERA
Bobby Jenks 1-1 1.08 ERA

If you want to buy scouting reports on players like these based on a little over 100 at-bats or so, we also have a bridge in Brooklyn that you may be interested in.

While winter stats are interesting they are not very useful. To give them any significant weight than another time period is foolish.

If a pitcher comes up with a new pitch, that is a different matter and that could mean that he has turned the corner and is ready to take off. However, that is fairly rare.

If you have a player who had a great winter season, now is probably a good time to trade him.

Return to Strategic Essay List
Return to PPI Home Page
Email questions to PPI or email support@stallvalue.com