Fun With Streaks and Slumps, Part 1
Tom Tippett
June 23, 1998
Sammy Sosa has been on an amazing homerun tear in the past month, with 17 homers in June (tying Willie Mays' NL record for a single month) and 21 homers in 22 games. And you could argue that Mark McGwire's 1998 campaign has been one never-ending streak.
Three years ago, Mike Benjamin set a major-league record with 14 hits (in 18 AB) in three games to raise his average to .447. According to Baseball Weekly, Benjamin had a .186 career average before this outburst. And before that 1995 season was in the books, his average had fallen all the way to .220, thanks to a 20-for-135 (.148) slump from June 19th to the end of the season.
Certain types of streaks get a lot of attention: winning and losing streaks by teams, consecutive wins and losses for pitchers, save conversions, hitting streaks, and errorless chances come to mind. But I'm going to focus on the tendency of batters to get hot and cold during the season. To gather the necessary, Luke Kraemer (a Diamond Mind software engineer) wrote a terrific little program that put every atbat from the 1997 season in chronological order, then determined the minimum and maximum number of hits and homers for any stretch of X atbats, where X ranged from 20 to 250.
Let's work through one example to see how this works. Barry Bonds had 532 atbats last year. That means there were 513 stretches of 20 atbats in the sequence -- 1 through 20, 2 through 21, and so on up to 513 through 532. In total, the program looked at 513 sets of 20 atbats, 512 sets of 21 atbats, and so on up to 283 sets of 250 atbats. If I remember my high school math well enough, that's a total of 91,655 sequences. The output tells us that Bonds' best stretch of 127 atbats produced 49 hits for a .386 batting average, while his worst sequence of 123 atbats saw him hit only .228, to pick just two of the more interesting runs in his season.
While Bonds' results were fairly typical of a player with his batting average, quite a few surprises showed up on lists of leaders and trailers. I'll present the batting average information in this article, and follow up with the homerun data in the next.
Hitless Atbats
It should come as no surprise that the longest hitless atbat streaks in 1997 were racked up by the guys who finished near the bottom in batting average. (Remember that this includes only those players with at least 250 atbats for the season). Here's the list:
Ordonez -- 0 for 37
CGoodwin, McRae -- 0 for 29
Paquette, Carter -- 0 for 28
GVaughn -- 0 for 27
Ausmus, Gant, BBoone -- 0 for 26
Sprague, Durham -- 0 for 25
Quite a few other players were equally impressive in their ability to avoid going oh-fer a long stretch. Wilton Guerrero's longest stretch of hitless atbats was eight. Mo Vaughn's been getting a lot of heat in Boston for striking out too much, but he never went more than nine atbats without a hit last year. And Bill Spiers was one of seven players (along with Joyner, Naehring, Orie, NPerez, Hollandsworth, and RAlomar) who never went longer than ten atbats without a hit in 1997. Even more impressive, it's been in the news this week that Tony Gwynn has never gone more than 14 atbats without a hit in his entire career.
Four players go 15-for-20
Sandy Alomar, Rusty Greer, Derek Jeter, and Kenny Lofton each had streaks of 15 hits in 20 atbats last year, demonstrating that Benjamin's 14-for-18 was most notable for being compressed into 3 games, two of which went into extra innings.
In fact, it's normal for a player to have at least one torrid stretch of 20 atbats in a season. Of the 268 players in our sample, only 10 failed to go 9-for-20 or better at least once. Three players (Ramirez, Mabry, WClark) had a 14-for-20 streak. Eighteen others were 13-for-20 at their hottest. Forty-two went 12-for-20, eight-five went 11-for-20, and another 63 batted .500 over 20 atbats at their peak. Chad Kreuter pulled up the rear with a high of 7-for-20.
Slumps of 20 atbats are equally common. Forty-six players had at least one 0-for-20, while 148 players bottomed out at 1-for-20. Only eleven players can claim that their worst stretch was 3-for-20.
Lofton bats .600 in 50 atbats
If we stretch things out to 50 atbats (or 2-3 weeks), we see that there are still quite a large number of very hot streaks. Led by Lofton's 30-for-50, seventeen players batted .500 for at least one series of 50 atbats. In all, 191 players had a stretch of 50 atbats where they hit .400 or better. And everyone in our group of 268 players had at least one run of .300 hitting that lasted 50 atbats.
The slumps were equally impressive. 238 players batted under .200 for at least one series of 50 atbats. The worst of them? Ed Sprague with two hits (.040), Ordonez with three, and a dozen players who went 4-for-50 (including Carlos Delgado). Of the 21 players who had a 5-for-50 slump, Cal Ripken, Mark McGwire, Hal Morris, Rafael Palmeiro, and John Valentin are the most notable.
Joey Cora?
The top 100-atbat streaks belong to Tony Gwynn and Joey Cora, who lit up opposing pitchers for a .490 batting average. Frank Thomas was next at 47-for-100, followed by Roberto Alomar and Mike Piazza with 46 hits each.
Naturally, at 100 atbats, fewer players were over .400 and .300 than at fifty. But it's still a pretty impressive number. Forty-three players hit .400 or better, while 242 of 268 hit .300 or better at their warmest.
The poster boy for 100-atbat slumps was Henry Rodriguez, who got coldest when the weather was hottest. His .176 average from June through August was due in part to a 12-for-100 slump. But he wasn't alone. Twenty-four players batted .150 or less over a 100-atbat run (including McGriff, Matt Williams, and Albert Belle). A total of 152 players had a stretch of 100 atbats at .200 or below.
The most consistent hitter over 100 atbats? None other than Tony Gwynn, whose worst stretch of 100 atbats produced a .290 average.
Leaders/trailers at 20-atbat intervals
Here are the top streaks of varying lengths...
Streak Average Players 15-for-20 .750 SAlomar, Greer, Jeter, Lofton 26-for-40 .650 Lofton 33-for-60 .550 Gwynn, Walker, Lofton 41-for-80 .513 Cora, Gwynn 49-for-100 .490 Cora, Gwynn 57-for-120 .475 FThomas 63-for-140 .450 FThomas, Gwynn 71-for-160 .444 Gwynn 78-for-180 .433 Gwynn 87-for-200 .435 Gwynn
and the worst slumps...
Slump Average Players 0-for-20 .000 46 players 1-for-40 .025 Sprague, Ordonez 3-for-60 .050 Sprague 7-for-80 .088 Sprague 12-for-100 .120 HRodriguez 16-for-120 .133 JKing, Sprague, HRodriguez 20-for-140 .143 HRodriguez, Brosius 24-for-160 .150 HRodriguez, AJones 28-for-180 .156 HRodriguez 31-for-200 .162 HRodriguez
Despite this horrendous slump, Henry Rodriguez finished the year at .244, thanks in large part to a 32-for-76 (.421) streak early in the season. Other notable slumps included Jeff King (34-for-198, .172), Todd Zeile (26-for-132, .197), John Valentin (9-for-83, .108), and Rey Ordonez (17-for-114, .149, and 31-for-189, .164).
Worst Slumps of the Batting Leaders
Even the batting leaders had their share of tough times. Here's a sample of some of their worst slumps:
Player Season Avg Slumps Gwynn .372 7-for-41 (.171) Walker .366 2-for-22 (.091), 8-for-49 (.163) Piazza .362 1-for-27 (.037), 5-for-40 (.125) FThomas .347 3-for-24 (.125), 9-for-51 (.176), 18-for-85 (.212) Lofton .333 2-for-29 (.069), 13-for-67 (.194) EMartinez .330 6-for-33 (.182), 13-for-57 (.228) Justice .329 3-for-27 (.111), 14-for-72 (.194) BWilliams .328 1-for-20 (.050), 4-for-34 (.118), 21-for-100 (.210) Ramirez .328 1-for-22 (.045), 9-for-46 (.196) Joyner .327 4-for-27 (.148), 9-for-45 (.200)
Conclusions
When streaks and slumps happen at the start of the season, they get noticed. When they happen later in the year, they often get buried in the season totals. I knew they were there, but I didn’t have a good feel for their frequency, magnitude and duration. I would not have guessed, for instance, that anyone (let alone Joey Cora) would stay hot enough for long enough to go 49 for 99.
What does this mean for fans and fantasy leaguers and real-life general managers? Stock market pros will tell you that you can't predict when the market will move up or down in the short to intermediate term. They'll also tell you that the long-term trend is up, so you want to be in the market. But the gains come in unpredictable bursts, and if you happen to be on the sidelines at the wrong time, you can give up all the rewards of being in the market for a given year. Their advice is to choose good stocks and mutual funds, then get in the market and stay in, so you'll be there whenever the upward surge appears.
While their advice may seem self-serving, since they make money when more people are in the market, I think there's a parallel with fantasy baseball. Choose good players, stick with them for the long term, and you'll have them on the roster when they heat up. Don't try to guess when they'll get hot, when a hot streak will end, or when a slump will suddenly reverse itself.
In the past few fantasy baseball seasons, I've tried to play the hot hand a few times, and most of the time I've been burned. Early this season I moved Devon White to the reserve roster because he started slowly. While he was on my bench for a few days, he went 15 for 28 (.536) with a homer, six runs batted in, and three steals.
I also caught lightning in a bottle with Sammy Sosa. He's a career .257 hitter who's been in the .330s all year, and his recent homer binge is unprecedented in baseball history. A day doesn't go by when I don't think about trading him for a pitcher because, after all, nobody can stay that hot for that long. And every day I thank my lucky stars that I didn't try to "time the market" and dump him at the peak.
The stock market analogy isn't perfect. Most investment pros recommend the market for folks who can ride it out for anywhere from five to thirty years. A baseball season is over in six months. So a patient, "buy-and-hold" strategy might not be the best approach, especially if you're trailing in the latter part of the season. But, most of the time, I think you'll be better off if you don't (a) panic when a key player goes into a tailspin, even for 50 or more atbats, or (b) get out too soon when someone gets hot.
Next Up, Homeruns
The homerun data was even more interesting, and I'll go into detail in my next article. For now, I'll leave you with a little quiz. You can probably look this up, but it'll be more fun if you try to guess first.
Q: As you all know, Mark McGwire slugged 58 homers last year. Care to guess how many homerless atbats he strung together in his worst drought?
A: Next time.
Copyright © 1998. Diamond Mind, Inc. All rights reserved.
