Analytics

JD08

Ready for Ska's fourth wave
Maude
Joined
Dec 19, 2014
Messages
35,181
In case anyone thinks it's a passing fad, it's actually gaining steam.

March 4-5: The inaugural Sport Ideas Symposium (sports-ideas.com/) is a two-day online event featuring three keynote speakers -- Brian Kopp (Phoenix Sports Partners), Alexandra Mandrycky (Director of Hockey Administration, Seattle Kraken) and Evan Wasch (Executive Vice President, Basketball Strategy & Analytics at NBA), as well as five panel discussions featuring industry experts from a variety of sports. The primary goal of the Symposium is to provide an accessible platform to learn about the dissemination and practice of analytics within sports organizations. Registration is only $10, and can be done at the link below
https://us02web.zoom.us/webinar/register/WN_-BnQPQbUQASxH6lt0GSD8g

March 18: The NFL's third annual Big Data Bowl will run live on YouTube, Twitch, and the NFL App at 3:00 PM EST. This event showcases 8 finalists (competing for an additional $10,000 in prize money), as well as a featured conversation with an NFL head coach known for his analytics use. The event is free to attend. Finalists for the event, as well as more information on the theme's covered this year, can be found at

 
Advertisement
In case anyone thinks it's a passing fad, it's actually gaining steam.

March 4-5: The inaugural Sport Ideas Symposium (sports-ideas.com/) is a two-day online event featuring three keynote speakers -- Brian Kopp (Phoenix Sports Partners), Alexandra Mandrycky (Director of Hockey Administration, Seattle Kraken) and Evan Wasch (Executive Vice President, Basketball Strategy & Analytics at NBA), as well as five panel discussions featuring industry experts from a variety of sports. The primary goal of the Symposium is to provide an accessible platform to learn about the dissemination and practice of analytics within sports organizations. Registration is only $10, and can be done at the link below
https://us02web.zoom.us/webinar/register/WN_-BnQPQbUQASxH6lt0GSD8g

March 18: The NFL's third annual Big Data Bowl will run live on YouTube, Twitch, and the NFL App at 3:00 PM EST. This event showcases 8 finalists (competing for an additional $10,000 in prize money), as well as a featured conversation with an NFL head coach known for his analytics use. The event is free to attend. Finalists for the event, as well as more information on the theme's covered this year, can be found at

LEAN is the present and future. The faster teams embrace it, the more competitive they will be.
 
In case anyone thinks it's a passing fad, it's actually gaining steam.

March 4-5: The inaugural Sport Ideas Symposium (sports-ideas.com/) is a two-day online event featuring three keynote speakers -- Brian Kopp (Phoenix Sports Partners), Alexandra Mandrycky (Director of Hockey Administration, Seattle Kraken) and Evan Wasch (Executive Vice President, Basketball Strategy & Analytics at NBA), as well as five panel discussions featuring industry experts from a variety of sports. The primary goal of the Symposium is to provide an accessible platform to learn about the dissemination and practice of analytics within sports organizations. Registration is only $10, and can be done at the link below
https://us02web.zoom.us/webinar/register/WN_-BnQPQbUQASxH6lt0GSD8g

March 18: The NFL's third annual Big Data Bowl will run live on YouTube, Twitch, and the NFL App at 3:00 PM EST. This event showcases 8 finalists (competing for an additional $10,000 in prize money), as well as a featured conversation with an NFL head coach known for his analytics use. The event is free to attend. Finalists for the event, as well as more information on the theme's covered this year, can be found at


My view since my Revenge of the Nerds day is that its always been a battle to the death between the jocks and nerds. This trend shows me that us nerds are continuing to win!!!!! And for you oldtimers, who can ever forget Miss Betty Childs!!!

1614718491563.png
 
Advertisement
Overall what is it. Don't kill me for asking.
If I had to state it simply:

A tool to make informed decisions and maximize your chances of success.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

No magic bullet in any form. If you are at a 36% probability to win and data allows you to make a decision to increase that to 38%, that's a win. You're still more likely than not to lose in this scenario, but making a better decision has improved your chances.

When the margins are as slim as they are in athletics (gap between each outcome), improving your percentages by fractions of a percentage numerous times throughout a season adds up to a big advantage. You aren't taking Kent State and making them Alabama simply through analytics. But you might take them from 3rd to 1st in their own conference etc.

Smart teams understand any advantage gained is positive EV (Expected Value). From there, analytics can help you decide WHICH positive EV project to take on. If a project gives 3% EV, but takes up tons of resources, it is preferable to take on a project that gives 2% EV but leaves enough resources to take on a second project that adds 2% EV and adds up to 4% EV with the same resources as the 3% project does.
 
Advertisement
Overall what is it. Don't kill me for asking.
Real world example:

If you're a head coach and you know that you're statistically more likely to do better going for it on 4th down at a particular spot on the field and a distance to go, would that affect your decision?
 
If I had to state it simply:

A tool to make informed decisions and maximize your chances of success.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

No magic bullet in any form. If you are at a 36% probability to win and data allows you to make a decision to increase that to 38%, that's a win. You're still more likely than not to lose in this scenario, but making a better decision has improved your chances.

When the margins are as slim as they are in athletics (gap between each outcome), improving your percentages by fractions of a percentage numerous times throughout a season adds up to a big advantage. You aren't taking Kent State and making them Alabama simply through analytics. But you might take them from 3rd to 1st in their own conference etc.

Smart teams understand any advantage gained is positive EV (Expected Value). From there, analytics can help you decide WHICH positive EV project to take on. If a project gives 3% EV, but takes up tons of resources, it is preferable to take on a project that gives 2% EV but leaves enough resources to take on a second project that adds 2% EV and adds up to 4% EV with the same resources as the 3% project does.
if analytics gives you only a 2% chance of winning than without, I would guess analytics isn't the main issue and you would be better spending your resources elsewhere like improving recruiting, coaching and player development..
 
Real world example:

If you're a head coach and you know that you're statistically more likely to do better going for it on 4th down at a particular spot on the field and a distance to go, would that affect your decision?
I'll do one that we all have seen, and that ties into Expected Value. It's 4th and 1 on the opponents goal line. You have a good kicker, who never misses from the 1 but your line does risk a block. Considering blocking concerns, you have a 99% chance of making the kick. Kicking has an EV of 2.97 (3 points × .99, the odds of converting).
You can also go for it, and let's say you have a 50% chance of converting 4th and 1. And let's pretend your kicker has a 99% chance of making the extra point, giving an extra point EV of .99. Since converting would be a touchdown and extra point, your EV is 3.495 ((6 + .99) x 0.5). So going for it is the better result in the long run if you convert at 50%.
In fact, in the above situation, given the EV of kicking, and the EV of a touchdown (6 points in the bag plus .99 since the extra point could get blocked), we can figure out how we'll we have to convert to justify going for it. If we do Kick EV/TD EV (2.97/6.99), we get 42.5%. If you convert 4th and 1 better than that rate, you go for it.

Obviously, keep in mind the goal is to win, and if it's the end of the game and all you need is a field goal, you take that. Even though it's lower EV in this case, it's the higher percentage play and it's good enough.
 
I'll do one that we all have seen, and that ties into Expected Value. It's 4th and 1 on the opponents goal line. You have a good kicker, who never misses from the 1 but your line does risk a block. Considering blocking concerns, you have a 99% chance of making the kick. Kicking has an EV of 2.97 (3 points × .99, the odds of converting).
You can also go for it, and let's say you have a 50% chance of converting 4th and 1. And let's pretend your kicker has a 99% chance of making the extra point, giving an extra point EV of .99. Since converting would be a touchdown and extra point, your EV is 3.495 ((6 + .99) x 0.5). So going for it is the better result in the long run if you convert at 50%.
In fact, in the above situation, given the EV of kicking, and the EV of a touchdown (6 points in the bag plus .99 since the extra point could get blocked), we can figure out how we'll we have to convert to justify going for it. If we do Kick EV/TD EV (2.97/6.99), we get 42.5%. If you convert 4th and 1 better than that rate, you go for it.

Obviously, keep in mind the goal is to win, and if it's the end of the game and all you need is a field goal, you take that. Even though it's lower EV in this case, it's the higher percentage play and it's good enough.

I'm a huge fan of Analytics - but it's always going to be tricky.

Example - is the 4th & 1 vs FAU, or is it vs Alabama?

On paper, obviously odds increase vs FAU, and decrease vs Alabama. But maybe in this particular game in short yardage situations, FAU has been stuffing you, and you've been having success against Alabama.

How much do you stick to/defer to Analytics vs reading the momentum and flow of the game?
 
Advertisement
if analytics gives you only a 2% chance of winning than without, I would guess analytics isn't the main issue and you would be better spending your resources elsewhere like improving recruiting, coaching and player development..
I think the biggest misconception is that analytics are something new that coaches are now deciding to look at to make decisions. Analytics have been around sports for decades at this point and coaches have used them to make decisions throughout the game, whether or not it’s implicit or explicit in the decision making process. To your question, who’s to say that analytics aren’t being used in recruiting and player development? They certainly are.
 
Obviously, no method is bullet proof but the anti-analytics crowd in sports is mind-boggling to me. Why would you purposefully not take every possible advantage to help you win? Successful businesses have use analytics for decades to help them maximize profits. There's not just some dude sitting in a board room making gut decisions about everything. They crunch as many numbers as possible to make the MOST informed decisions.
 
Advertisement
I think the biggest misconception is that analytics are something new that coaches are now deciding to look at to make decisions. Analytics have been around sports for decades at this point and coaches have used them to make decisions throughout the game, whether or not it’s implicit or explicit in the decision making process. To your question, who’s to say that analytics aren’t being used in recruiting and player development? They certainly are.

Yes... but I was replying specifically to the example provided of winning and losing. Improving your chances of winning by 2% seems pretty negligible to me.
 
Yes... but I was replying specifically to the example provided of winning and losing. Improving your chances of winning by 2% seems pretty negligible to me.
It wouldn't if your chances were 49%.
 
Yes... but I was replying specifically to the example provided of winning and losing. Improving your chances of winning by 2% seems pretty negligible to me.
In a game of inches with millions in salaries and revenues on the line, most head coaches around the country clearly disagree with that.
 
Advertisement
Back
Top