As a data analytics professional (my day job) I have to chime in. I suspect that analytics as a tool didn't fail here. What failed was that it appears Tampa Bay froze their analytics 'model' at the end of the regular season and did not update their model (or their model did not react quickly enough) based on new (more recent, and more relevant) data from the post season (i.e. that Nick Anderson had been struggling lately).
This doesn't make a whole lot of sense. I think if I'm making a decision like that based on one moment then I am doing myself a disservice.
"When it matters most" isn't a thing. Every right decision on the road to get to that moment mattered just as much.
What letter grade would you give this six game series?
I would absolutely expect players to always want to be on the field, and I totally understand Snell's frustration. I personally wouldn't make a career choice based off getting pulled an inning early in a couple of games though, especially if I want to win. The team was two wins from a title.Uhhhh.... we're going to have to agree to disagree here. If you don't think that top athletes care about being 'the guy' in the big moment when they are the star of their team then I think there's a bit of a disconnect. If Snell happily gave up the ball and said 'Sure thing, Kevin. I trust you. Maybe I can't get these guys out.'... that's probably not a player you want on your team, especially in crucial moments.
I don't know how you could think Snell isn't absolutely fuming and pissed off with being taken out there. Regardless of the 'right decisions' that were made in the regular season.
I would absolutely expect players to always want to be on the field, and I totally understand Snell's frustration. I personally wouldn't make a career choice based off getting pulled an inning early in a couple of games though, especially if I want to win. The team was two wins from a title.
I think it's strange to dump on strategic thinking that helped get a team so close to a title. Unless you think cash was holding Tampa back that entire time. The rest is just hyperbole.But the fact that it wasn't just a one off 'gut feeling' move sets the precedent now. Snell has to know that unless the team does a 180 on their philosophy, they're going to make big game decisions based on spreadsheets rather than how players are actually performing on the field. Not saying that you shouldn't use analytics of sabrestats to guide your moves but at a certain point you have to watch what's happening in front of you and what the situation is. Can't see how an ace who wants to win is OK with that.
"Coach, how can you take me out? I'm pitching a perfect game!"
"Yeah... but 3rd time through the order..."
"Right, right. My bad. Here's the ball."
Years Waiting for World Series Championship:
0 - Dodgers
1 - Nationals
2 - Red Sox
3 - Astros
4 - Cubs
5 - Royals
6 - Giants
9 - Cardinals
11 - Yankees
12 - Phillies
15 - White Sox
17 - Marlins
18 - Angels
19 - Diamondbacks
22 - Rays
25 - Braves
27 - Blue Jays, Rockies
29 - Twins
30 - Reds
31 - Athletics
34 - Mets
36 - Tigers
37 - Orioles
41 - Pirates
43 - Mariners
51 - Brewers, Padres, Montreal
59 - Rangers
72 - Indians
136 - Providence
I think it's strange to dump on strategic thinking that helped get a team so close to a title. Unless you think cash was holding Tampa back that entire time. The rest is just hyperbole.
The silver lining in this cloud is that hopefully it puts to bed the notion that Clayton Kershaw can't pitch in the playoffs.
One season doesn't erase his postseason failures imo. And frankly I'm not sure one title totally erases the Dodgers failures per say. One championship in nearly a decade of dominance. You hope to see more titles.
I consider the entire idea of certain players or teams being "playoff chokers" to be utter nonsense in the first place. Hopefully this just results in fewer people repeating it.
I think he made the mistake of relying too much on one set of data.5.1 innings
2 hits
9 K
0 BB
73 pitches
taking him out there is not defendable by any means regardless of what the "advanced stats" say. When your freaking ace is dealing like that you sit the F down and let them go and to actually take him out because a probability spreadsheet says to is the definition of stupid.
if you're going to manage strictly by what the numbers say to do then why even have a manager? just have a computer spit out the decision each pitch and play the game like an EA Sports simulation
I think it's strange to dump on strategic thinking that helped get a team so close to a title. Unless you think cash was holding Tampa back that entire time. The rest is just hyperbole.
I agree with most of this. Entertaining only one line of thought in any situation is asking for trouble, and that is likely what happened in that situation.I think there's a time and a place for sticking to your analytics driven game plan. Like I said earlier, it's an approach that provides the best results over a large sample size. So, essentially the regular season. You don't have that luxury in the playoffs. It's one off high pressure situations where you need to give yourself the best opportunity to win in that moment.
Say it's game 7. Bottom of the 9th of a tie game with a runner on 3rd with 2 outs and Cody Bellinger is up at the plate. Are you going to do some sort of extreme shift on him because that's what the stats say is the best approach to get him out? Like, I guess you can but then you leave the entire left side of the infield open and he could bunt the ball to 3rd base and you lose the series.
Yes, these examples are hyperbole but I'm using them to make a point. You have to take the stats and analytics and be constantly adjusting them in game based on the situation. YOU CAN'T JUST SAY 'I'M TAKING SNELL OUT 3RD TIME THROUGH THE ORDER BECAUSE THE NUMBERS TOLD ME TO. Maybe if the guy you're putting in is lights out and is on some unnhittable streak... but that's not what Cash and the Rays did. They said 'Well, this worked in spring training and Andersen had good overall numbers so it should work out now in this one isolated incident'. That's the problem.
When you win you need to win big, when you lose, make it small. If you just treat every situation like it's the exact same then you might win more often than not but you'll have no control over the highs and lows. You can't win in the playoffs like that.
Cash deserves some blame. But what about the dude giving up the lead. Why absolve him of any blame? He could've made the Snell decision a footnote.
I also don't think they would fire Cash. He was a huge part in them getting here.
I think he made the mistake of relying too much on one set of data.
Honestly, if they did replace managers with computers for lineup decisions teams would probably perform better. The value in managers lies in managing personalities and putting players in the best position to be successful when they are eventually put on the field.
Yeah, my issue isn't that Cash pulled Snell that early. It's that he replaced Snell with Anderson, who's been badly struggling the entire post-season. Relievers are strange beings, and if one has been struggling as bad as Anderson has been throughout the post-season, you just can't trust him in this high-leverage situation.
I agree with most of this. Entertaining only one line of thought in any situation is asking for trouble, and that is likely what happened in that situation.
However, the extreme reaction to that decision is based on the result, not the process, and I think that is just as dangerous. Baseball is unique in that there is much more randomness built into the game, far more than any other sport, and that gives managerial decisions the appearance of a false degree of weight.
You're assuming that a computer can't do those things, but a human being can? There are obvious "intangibles", but I don't think they can be any more accurately ascertained by humans than computers.It would be a disaster to have a computer as a manager because there are variables you can't quantify. Like if a player is maybe at 80% capacity on a given day there's no way of really knowing that, and there's no way to know how well they're actually going to perform at that level. Whether that's as a pitcher, a runner, a hitter, an outfielder it doesn't matter. You have umps with inconsistent strike zones as well to deal with. A computer is going to spit out decisions based on data that's in a vacuum and the game isn't played that way. Mostly a computer would probably explode when it came to bullpen decisions because it would look at some pitchers with a tiny sample size an think they're the re-incarnation of Bob Gibson and run them into the ground and they'll eventually give up a billion hits and runs and flame out. It also would probably replace every catcher with a player who is a better hitter because catching metrics are far from an exact science in baseball so now you have a bunch of better hitting, but terrible defensive catchers. So now pitchers are going to suffer which will further mess up data sets even further.