No video :(

SSAC16: Winning at Daily Fantasy Sports Using Analytics

Подписаться 12 тыс.

Просмотров 11 тыс.

50% 1

We present an analytics based approach to winning daily fantasy sports hockey contests which have top heavy payoff structures (i.e. most of the winnings go to the top ranked entries). Our approach incorporates publicly available predictions on player and team performance into an integer program that computes optimal lineups. Using our algorithm, we were able to place in the top three in contests with thousands of participants multiple times. Our approach can easily be extended to other sports besides hockey, such as American football and baseball. Come learn how you can use our algorithm to beat the system in DFS!

Опубликовано:

5 сен 2024

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 9

@doofus9575 5 лет назад

The lesson this should teach you is that so many things in life are all about intelligence. These guys, from MIT of course, knew nothing about hockey coming in, and they made an enormous profit. Someone else may know all there is to know about hockey, and not be able to bet worth a damn, because they're not bright enough to put aside their own emotions and figure out what matters and what doesn't.

@ManMonkey600 7 лет назад

This is a great video and presentation.

@lillyscott9333 7 лет назад

What is the keyword to use at github to download the code used?

@Excuse_The_Hebrew 7 лет назад

no website?

@stevenorthenscold6313 6 лет назад

How do I use this info in the NFL?

@joshschreder 7 лет назад

Where can I find the optimizer?

@ManMonkey600 7 лет назад

Anyone can build or use an optimizer. It is really the projections you feed into the optimizer that are the most important part of creating multiple lineups. Exactly what they say is true. The more lineups you enter the less profitable you will be since you drift farther away from creating optimal lineups. You can even use Excel to do a very basic solver formula.

@ChowChow414 6 лет назад

I am using an approach similar to this. But I actually build the sparse co-variance matrix for MLB, and use a Quadratic program to solve it. The covariance matrix figures out who to stack and pick.

@delt19 5 лет назад

@@ChowChow414 I found your GitHub and saw that you're working on an NBA model. Is that still in progress? I've been building several models in R for FD NBA and would be interested in discussing some ideas if you're still building your model.