UPDATE (6;29 p.m.): For Adam and Steven T. Truth, the answers to your questions are at the bottom of the post
Still plenty of debate out in the blogosphere about whether the Mariners are actually close enough to the post-season to be gambling anything on a playoff bid in 2008. I’ve seen some comments from folks wondering whether M’s general manager Bill Bavasi even knows what a Pythagorean expectation is all about. I can assure you that he does. Does he consider it the be-all, end-all? I sure hope not. First off, there’s a margin of error of roughly four games in either direction when it comes to a Pythag record. So, the Mariners, expected to “only” win 79 games last year based on their runs for/runs against differential, could have won anywhere from 75 to 83 and still fit well within the statistically accepted norms.
They did win 88, so there’s something else going on. Many of you have mentioned the bullpen helping the team win “higher leverage” games. In fact, a study done last year by David Gassko in The Hardball Times found that teams with balanced offensive lineups and strong back ends of bullpens tended to outperform their expected win totals more than other clubs. Gassko threw in extra points for strong managers as well, though not to the degree as for the two other categories. One can debate the merits of Seattle’s dugout management, but much of the raving about last year’s team centered around the bullpen and the more-balanced offensive attack, which saw all nine regulars collect 50 RBI or more.
The crux of the article is that there are still factors outside the realm of statistical measurement that can impact a won-lost record. In other words, we can’t measure everything just yet. Something to keep in mind. I personally believe the Mariners were more in-line with an 83-win team. But maybe it’s really an 85-win team. Is that enough to take a run at the post-season by adding an ace? Exactly how good does a team have to be from year-to-year to have a shot at the playoffs? Well, I compiled some won-lost data from all playoff teams starting in 2000. Some of the results might surprise you. Namely, the idea that you need to be at least a 90-win club one year to have a shot at the playoffs the next.
In fact, of the 64 major league playoff teams from 2000 to 2007, the average win total they had posted the previous season was 89. Yes, that would be one win more than last year’s Mariners. One single victory. So, if Jeff Weaver doesn’t blow a 5-0 lead at home to the Angels last August, the M’s would have posted exactly the number of wins that this decade’s average MLB playoff team did in the season before they actually played in the post-season.
Now, look. Of course there was some overlap in the math. Teams like the Red Sox, Yankees, Cardinals and Braves obviously have made the playoffs multiple times consecutively over the years. So, the year before one of their playoff appearances could actually have been a playoff season for them as well. Meaning, a good season with a high win total.
But if anything, that would tend to skew the win totals to much higher lengths.
I mean, we could look only at playoff teams that didn’t make the post-season the previous year and see what their win totals were. It would flatter the M’s, I’m sure. So, let’s try it.
By my calculations, there are 29 playoff teams you can measure since 2000 that were not repeating a post-season performance from the previous year. In other words, they went from also-rans, like the current M’s, to playoff prosperity the next year. So, what was the average number of wins for those 29 teams the year before they went to the post-season?
Yep. You read it right. Those teams went from an average of 80 wins one year to playoff fruit the next. Sure, some of them were rebuilding and bound to hit the jackpot big-time. Some were playoff teams that took a year off because of whatever and found themselves back in the post-season the next. But not all 29 of them.
That’s 80 actual wins recorded on the actual playing field, not in some computer’s prediction. The Mariners, let’s not forget, won 88 games last season. So, in real life, they are well ahead of the curve when it comes to being “qualified” to make a playoff leap.
OK, then, what about the big Pythag difference we talked about? I don’t want to minimize this, because I do think the M’s overachieved last season and were not a true high-80s win team. More a low 80s, at best a mid-80s. How much does that matter? Let’s see.
The average expected Pythagorean win total of those 29 teams the year before they went to the post-season? Well, it’s a little stronger. A grand total of 83 wins, to be exact. So, that’s four wins better than the 79 that the Pythag formula says the M’s “should” have had last year.
But wait! What about the margin for error? You know, that pesky four-win differential either way? Ah, yes. The one that says the M’s could drop as low as 75 wins, or as high as 83, and still fit squarely within the formula’s parameters? Well, let’s see, that high of 83? Matches the average Pythagorean win total of those 29 teams the year before they went to the post-season. In other words, the M’s are right on-target to make such a leap.
I know this is a lot of numbers to swallow. But either way, either with real wins or computer projected wins, the Mariners are in the same boat as those other teams that jumped from out of the playoffs to into the post-season within one year.
Are there problems with this math? Of course. If you break it down into AL or NL totals, the numbers fall slightly out of Seattle’s favor. There is less chance of an AL team “breaking through” to the post-season nowadays because of so many higher-priced behemoths in their way.
Let’s split it up.
Since 2000, all 32 AL teams to make the playoffs averaged 90 wins the previous year — both in actual victories and their expected Pythag total. In the NL, it was 89 actual wins and 88 on the Pythag scale. So, slightly tougher in the AL. The M’s had 88 actual wins, so they are within striking distance of the teams in both leagues that went on to the playoffs the following season. It’s the Pythag area where the gap looks tremendous — 79 for the M’s and 90 for the other AL teams that went to the post-season one year later. But when we throw that margin of error of +/-4 in there, the gap quickly closes to as many as 83 wins for the M’s and as low as 86 for the other AL teams.
Not as formidable, but still a gap. This is where the first part of this post comes in, along with the variables — like a strong bullpen and offensive balance — that Gassko talked about. Could these be enough to close the gap for Seattle? We simply do not know with any degree of certainty.
When it comes to non-playoff teams jumping to the post-season in one year, the AL is even that much tougher than the NL.
Of the 29 clubs overall to have made this jump since 2000, only 13 came from the AL. And they had an average win total of 82 wins the previous year, compared to only 79 for the NL teams making the leap. So, the AL teams were slightly above .500 before taking their playoff shot.
So, a little tougher, though it must be stated again, the M’s won 88 last year. What about on the Pythagorean scale? The expected win totals of the AL and NL teams before making that post-season leap? Funny enough, it was 83 wins on both counts. Once again, even with a Pythag expectation of 79 wins last year, the M’s still fall within the range of an 83-win team because of the +/-4 win margin for error in the formula.
Does this prove that trading for Erik Bedard is the best move to make? No, it does not. What it does show, is that arguments the Mariners should not trade for Bedard, based on their Pythagorean record from last year, may be off-base. The numbers I’ve shown you demonstrate that it is not out of the question for a team like the M’s to expect to make a leap from second-place to the post-season within one year by vastly improving the starting rotation.
And remember, lets not forget the intangibles like the bullpen and a more-balanced offensive attack, that can lead to wide varitations between a Pythag prediction and an actual won-lost record. Based on what I’ve shown here, coupled with the potential for a bullpen/balance-related win total swing, I don’t see how folks can say Seattle is too far behind to at least try to make a playoff leap.
I have not been the biggest fan of how this team has been run since I arrived on the scene in September 2006. But if the club can add Bedard, I will be impressed by what it’s done this off-season to close the gap with other divisional and wild-card rivals. And from what I’ve looked at, the Pythagorean win total the Mariners posted last season is not a good enough reason to justify dismissing an Adam Jones/George Sherrill/plus trade for Bedard. If you want to argue the prospects package with me, then fine. But arguing that Seattle isn’t close enough to the playoff field to even take a shot? Not buying it. Neither, I suspect, would Pythagoras.
UPDATE (5:48 p.m.): For Adam in the comments thread, here is the list of teams, both AL and NL, that you asked for…
(previous year’s win total in brackets, followed by Pythag total from previous year)
White Sox (75…72)
Red Sox (93…100)
White Sox (83…84)
Red Sox (86…81)
OK, now on to Steven T. Truth’s question about how many teams have out-performed their Pythag expectation by at least five games one year, then made the playoffs the next. I found 15 of them since 2000, so it’s a lot more common than you think. Here is the list (actual wins from previous year, followed by Pythag total from previous year, in brackets).
Red Sox (86…81)