<div class=”d4p-bbt-quote-title”>biggill wrote:</div>
This is exactly why analytics is going to grow in hockey. Why did they win the cup? The team that can decode that mystery and show trends that improve odds is going to succeed.
Analytics describe what happened in the past. They can indicate what’s possible, but they don’t tell you how to make it happen again in the future. It’s steering the boat by watching your wake.
They also don’t show the ultimate potential. What do the analytics tell you about Kaprisov, is he a 20 goal guy when he gets to the NHL or a 40 goal guy?
Sorry, but I don’t have the faith in past numbers as an indicator of future results. Look at baseball, there’s a stat for everything so how come nobody’s cracked the code and winning 90% of their games?
The analytics on the Blues last season will tell you that they won the cup because they got hot goaltending and totally awesome play from O’Riley at the right time. I’m sure glad we have a supercomputer to crunch those numbers and give us the real story of what happened.
Grouse
I think there might be a bit of confusion around what analytics is, which is understandable because it’s an extremely broad and ill-defined term. What you are talking about is actually what I would call reporting. Normally the least mature type of ‘analytics’, it’s describing what happened in the past.
There’s an entire category of analytics devoted to making predictions for the future(predictive modeling), that is more or less the point of analytics in many cases.
Then, there’s an even more advanced category of analytics called machine learning, which is like a predictive modeling system that continues to improve and learn as new data is constantly fed to it(historically models were time consuming to build and develop, so they would be built on a static set of historical data, over time this would lead to the model becoming ‘stale’ if it didn’t take new data into account. Think of predicting sales for a retail store, but we get hit by an unpredictable depression midway through the year, this impacts our future predictions but wasn’t accounted for in our initial predictions.).
On top of that, we have AI, which is often taking Machine learning type systems and using them to automatically make decisions. Take self-driving cars, they are trained models that tell the car what to do, when to stop, how fast to go, etc. That system is on the fly making decisions to slow down, speed up, turn, stop for that pedestrian(or not stop for kangaroos in this case
). Those ‘models’ are constantly being fed enormous amounts of new data to make them smarter and adapt to new conditions/situations.
It is not steering the boat while watching your wake.
They will figure out next-gen tech and stats for hockey, look at what they’ve done in the NFL to better understand separation between WR and CBs, or win/loss type metrics for linemen. It’s going to happen, and the teams that figure it out first will gain a bit of an edge in identifying under-rated players, exploiting weaknesses/tendencies of opponents, etc.
It’s like the silly argument that the shift in baseball doesn’t work because I saw a guy get a hit one time that normally wouldn’t be. Well if shifting saves you 20 runs over the course of the season, that’s a net win. You’re not going to win a world series because you saved 20 runs, but it’s a piece of the puzzle, finding those little margins.
Saying something doesn’t work because one team hasn’t figured out how to literally win every single time is just a weird way to look at it. It’s a piece of the puzzle, and the teams that figure out this piece will give themselves a bit of an edge over the other teams, given they are competitive in the other aspects of running a team. And the more pieces they figure out, the bigger that edge becomes.
As for teams cracking the code? Well, once one team innovates and figures something out, the other teams follow-suit. You are looking for little things to exploit that will likely only exist for a couple of years. And it 100% happened and is happening in MLB, so that’s a terrible example to use for your argument.
Now, if all you’re saying is you don’t like it or how it’s changed MLB, that’s fine. That’s a perfectly acceptable opinion, but saying it doesn’t work or hasn’t provided value for MLB teams…not a great argument.