Nationals Baseball: All In or Not All In, that is the question

Thursday, November 13, 2014

All In or Not All In, that is the question

Do we try to win now, or do we try to win later?

It's the eternal question for a baseball team. Do you spend money in the years directly infront of you attempting to maximize your potential of winning it all right now, or do you defray costs in order to maximize your potential of having a consistent playoff team later.

For various reasons I'm usually a win-now guy. The idea that you have to choose is founded in the "money bucket" theory that you guys know I hate. Also the future in sports, even 3 years down the road, is too variable to plan on. But let's try to take a more mathematical approach.

Let's say you make the playoffs and you have a 50% chance of winning any series. That's really rough but also pretty fair.  Your chances of winning it all in this purely mathematical scenario would be only 12.5%. If you made the playoffs with these odds you wouldn't expect to win a World Series until you made the playoffs 6 times.

How does that work? Well if you have a 12.5% chance of winning, you have an 87.5% chance of not winning. Your chances of not winning in back to back years is 87.5% * 87.5% or 76.5%.  This continues on and on and your chances of not winning 5 years in a row is still above 50%. (51.3% to be precise). So, if you made the playoffs 5 years in a row and had a 50% chance of winning every series, you'd STILL be more likely not to have won a WS than to have won it. Just barely but it's so. In the 6th year it flips.

OK so what if you go all in? How does that effect things? Well depends on how much an advantage you think it gives. Some people jump right away and say something like... 75% chance of winning a series. That's ludicrous. The best teams ever playing a mix of playoff teams and terrible teams don't hit a 75% winning percentage. 75% is way too high.  Let's look at it another way. The Nats won 59.3% of their games playing a mix of teams. How many games would they win playing ONLY playoff teams? Hopefully you can see that even 55% is generous. (in fact the Nats played 46% ball versus playoff teams this year, but 56% versus NL playoff teams if you want to parse further)

Let's go with that for now though. What changes if the Nats have a 55% chance of winning every series? Their chances of winning skyrocket all the way to... 16.6%, which makes their chances of not winning 83.4%. Not a big change right? BUT there is a big difference if you look over the course of years. Instead of hitting the "should have won" point in that 6th year, you now hit it in the 4th year. By that 4th year your chances of having won at least one world series are 10 percentage points better. ~52% rather than ~42%

And we're comparing it to an even-steven team. If the "All-in" Nats need 4 years to win, what would the "just get in" Nats need? Well if all-in wins at 55% than just in would win around 45% right (once again forgive the roughness)? They are the worst in the playoffs.  Run the numbers... carry the one... these Nats would need 8 years to get to that same "should have won one" point and in comparison to the all-in Nats in year 4 have a chance to have won at least one World Series that's 20 percentage points lower.

So going all-in as opposed to just sneaking in makes a big difference. It may not seem so initially but those little percentage differences add up. Of course there are two big issues here, or maybe three, or even four, now that I think about it. The first is the 55/45 split is probably too generous. 53/47 is probably more in line. That decreases that year 4 difference to 12 percentage points from 20. Still big but almost cut in half. The second is the window sizes we're talking about. Is an "all-in" window of 4 years viable? A "just make it in" window of 6+? Probably not. When we're talking "all-in" we're usually talking about a year or two, versus 3-5 years if you try to just be good enough. Based on the numbers and the 53/47 split you'd (juuuuuuuuuuust barely) rather have 3 years of "good enough" over two years of "all-in". Even if that's a wash, 4 years of making it with what are considered bad playoff odds beats 2 years of making it with good ones.

So it seems like "good enough", when applied to the reality (well somewhat - this is all still just literally numbers on a spreadsheet), makes the most sense. But then you remember - I'm not factoring in the WC.  The "all-in" team should avoid that. The "good enough" team would possibly have to face that at least once, probably a couple times.  Toss that in twice in four years and "all-in" for 2 takes the juuuuuuuuuust barely lead. Plus I haven't factored in the chances of winning MORE than one.

But it all leads to a single question really - what matters to you. Going all in for a couple of years will very likely slightly increase your chances of winning a title. But it's slightly. We're still saying your chances of NOT winning one are close to 70-75% instead of  80+%.  Is that enough of an improvement to you if your team is knowingly going to fall out of contention afterwards?

For a team about to tumble into the void regardless it makes sense to go all-in. For a team that holds out hope of another window opening soon the question is valid just from a competition standpoint.  The all-in advantage for winning one series is limited and decreases over time. Business wise "all-in" is not really a sensible idea. The money you'd save by not doing it and presumably make from more playoff appearances and regular season wins probably offsets the small increase in chance that you win it all. You can see why teams avoid the all-in more often than not.

There isn't a big enough difference here to say one way is obviously superior to another. There is an argument for either side based on what you'd expect to see on the field. All-in or not all-in doesn't have a factual answer, it has your opinion. Me? I'm all-in. 


Donald said...

Okay. Here's my issue with this. Saying "Go all in" sounds good, but what does it really mean? Does extending Znn, Fister and Desi mean the Nats are all in for the next few years? Or does not extending them and using the money for someone like Sandoval mean they are all in for 2015?

Is being all-in defined by having a large payroll? It doesn't seem so or we'd say the Phillies were all in last year. So maybe it's having a large bump in your payroll? But if that's the case, would the Astro's be all in by adding $20m next year and would that really increase their odds of winning the World Series? Or is it defined by trading away top prospects for ML ready players? That might be the case if you can only be all in for 1-2 years but then the risk / reward goes down. Or can the Nats be all in just by NOT trading Znn, Fister or Desi this off season? Can the Nats decide their odds of making the playoffs are good enough for 2015 that they can trade Znn now to significantly increase their odds for 2016-2018? Or does being all in imply that you also have to be making large sacrifices in out years?

Were the A's demonstrating 'all-in-ness' last year, or just making panic moves? Are the Dodger's all in for 2015 or do they just have a bigger payroll than smaller market teams?

Harper said...

Donald - See! No clear facts! It is what you make of it. Trippy music and psychodelic lights!

I'd define "All-in" as an attempt to address any issue you have with the best or near best available solution regardless of cost in $ or prospects (though prospect deals should be close to fair). From there you have to look at the time frame you are interested in. For the Nats all-in in 2015 simply means going after a 2B (in trade because FA options are unappetizing) and potentially another bullpen arm and not trading away anyone. For 2016, it would meaning keeping Ian, ZNN and Fister (or signing their equivalents)

Payroll size and increases are usually associated with "all-in" but not necessary. "Can the Nats decide their odds of making the playoffs are good enough for 2015 that they can trade Znn now to significantly increase their odds for 2016-2018?" is not really an "all-in" question. You're not setting up an all-in scenario at any single year.

A's were leaning toward all-in last year (actually were trying to maximize 2014 PLAYOFF win chances - almost cost them getting in and then Lester f'd it up) Dodgers would be all-in if they re-sign Hanley, sign Russel Martin and work on the pen

Lee said...

In all honesty, I feel like the Nats can go with the "good enough" approach for 2015 and still win the East comfortably. We don't really need to make any big splashes this off-season. Save for LaRoche, the team that just finished in first place is largely in tact. In 2015, the Mets should be better and the Marlins could be better, but the Phils will likely be worse if they sell off parts and I imagine the same for the Braves.

It's 2016 and beyond where we should learn what kind of team this will be over the long haul. Rizzo might make some moves this off-season that reveal an all-in or "good enough" inclination (or possibly the all-in forever approach that eventually burns out, a la Yanks), but I doubt it will happen. The decisions will be made only when they need to be made. The only all-in move we've really seen out of Rizzo was the Soriano signing. And really, was that his move?

I'd bet, that short of shoring up 2B (or 3B) for 2015, no big moves or extensions means the Nats are all-in for 2015.

Q: How does one perhaps get the most out of Desi, Znn, and Fister?

A: Let them play for contract $$$$$.

John C. said...

"All in!" is cool. It's macho. It's interesting, because it automatically raises the stakes - which makes the outcome that much more riveting. It's also (as Donald pointed out) pretty squishy in terms of being a guiding principle.

And, as Harper noted while declaring himself an "all-in" guy, it's not really smart/sensible even if one accepts the yanked-out-of-an-orifice percentages. Which I question - I'll note that over the past 15 years the 30 #1 seeded teams made the WS less than a straight 50% would have indicated (only 7 of 30 teams made it, with 15 of the 30 teams losing in the DS and 8 more in the CS round). So making your team the best overall team doesn't really seem to help you get to the desired championship, it just gets you a coequal seat at the table.

And redefining "all in" to being "best positioned to win it once you're in it" doesn't really work either. A recent Fangraphs article crunched some numbers and concluded that there aren't really any specific team attributes that strongly correlate with winning in the playoffs. The important thing is to be good enough to make it.

Chinatown Express said...

Harper, I take issue with your math. It's certainly true that "[t]he best teams ever playing a mix of playoff teams and terrible teams don't hit a 75% winning percentage." But you don't need a 75% winning percentage to win 75% of your series. Because a 7-game series is a larger sample than a 5-game series, a small edge game-over-game can result in a much larger advantage series-over-series. A thought experiment: If the playoffs were a 162-game series between two teams, a team with an expected 55% winning percentage would be a strong favorite to win the series. And if you played multiple series, you'd expect the 5% favorite to win nearly all of them, even if it's only winning 5% more than half the games. The same thing applies, on a much lesser level, to a 5-game or 7-game series.

As a separate issue, related to your point about the wild card issue, your all-in team also has a much better chance of winning homefield advantage in the DS and CS, which must be worth something, right?

KW said...

Billy Beane, tired of the playoffs being a "crapshoot," went all in last year, grabbed everyone he could get, and so unbalanced a team that at one point was the best in baseball that it almost failed to make the postseason and then exited meekly. The Phillies went all in a few years ago to re-sign everyone in sight (except Werth, the only one who has actually aged well) and have been stuck with a decaying, overpriced, noncompetitive roster. The Braves had the fad last offseason to overpay their kids, only to fall apart, fire their GM, and now talk of not contending again until 2017. So . . .

You've got to turn over your roster. You've got to plan ahead. You've got to look long and hard at the wall Rollins and Utley--both BETTER than Desmond (and Hanley Ramirez)--hit in their 30s. You take seriously the info on the pitchers with $100M contracts.

If it's me, I'm probably only really trying to re-sign Fister. I'm definitely trading Clippard this offseason and dangling Span. I'm probably riding Desmond through 2015, in large part because we've got no other options. And I'm doing exactly what Rizzo is doing with J-Zim: letting it leak that he'd be available for the right price, then denying it like crazy. But I'd be looking for Baez as the starting point of my return, unless they wanted to make it something like Castro and Russell. (But c'mon, a guy named Addison HAS to play at Wrigley, right?)

In short, I'm planning ahead to be somewhat all in for the next five-plus years, not just 2015.

akın sürmeneli said...

10 yılı aşkın bir süredir ümraniye dudulluda sex shop ve erotik shop sektöründe faliyet gösteren firmamızda dünyanın en ileri gelen markalarının seks shop ürünlerini güvenle alabilir ve kullanabilirsiniz
yurtiçi ve sürat kargo ile teslim alırken kapıda ödeme yapabilirsiniz, isterseniz kargolarınızı ikamet adresinizden yada seçtiğiniz kargo firmasının size en yakın şubesinden teslim alabilirsiniz, maximum kartlara kapıda 6 taksit yapabiliyoruz