Hi all,

New to DFS but looking forward to learning it profitably. I hear adam on the 2+2 pokercast talk about how its pretty impossible to beat, but im wondering what makes these pros so good? Is it the software they use for modelling lineups?

which software should I look to learn and what are the techniques that make DFS profitable?

EDIT: most, if not all, of my play will be on golf, so looking for modelling and/or strategies on PGA

DFS software is like feeding a baby--introduce things little-by-little, in order to be able to identify an allergy.

FantasyLabs is great, but if you're just diving in, it can be overwhelming. Excel usage is #1 and there aren't really many shortcuts to that. Even with such subscriptions, you have to do some of your own work to put the mess together.

I don't play PGA, but finding your own process of putting all players in one place with the data you need (evidence of the things which create fantasy points) is best done with a spreadsheet where you import the data.

Ownership is huge in PGA, and I've found the best way to be contrarian in other sports is to listen to podcasts and red articles, noting the mentions in my sheet, but if you're doing your own research on your own sheet, isolating your chosen data, you will have basic elements of contrarian play there already.

Iphone 6.

Sometimes I read an article or two.

Serious question, has he ever had a pro on the podcast or is he talking from his own perspective?

A lot of it is digging into research and there are a lot of places you can go to find the information like Labs, Rotogrinders etc.

I'd imagine, at the very least, Max Steinberg has appeared on the cast, siege.

anyway, yeah, fantasy labs and RG are two good resources. Labs is cool because they allow you to essentially build your own models.

Only other advice is to dig as deeply as you can and identify the truly predictive data, which both FL and RG can help you with.

Lol fantasylabs. Their head "stats" guy said something on twitter like "we don't provide actual points projections because they are too noisy." When you see a claim like that, you should run, not walk away.

I know he has had Rayofhope and AEjones on. But I still feel he is obviously talking from his own perspective.

LOL points projections

LOL at calling Bales the "head stats guy" and apparently not knowing the meaning of noise.

in any case, fantasylabs is a useful tool; the trends tool allows you to test out what you think you find in your research, and you can build your own model (based on data and value) and backtest it to get an R correlation.

but i should ask, what do you mean by software op? are you asking about optimizers?or just about the best sites to do research (my vote goes to fangraphs hands down), or what ppl use (as stated, excel or sheets, or both).

there isn't anything like a HUD in dfs.

It's wasn't Bales. I can't find the original tweet but luckily they posted something very similar on the FL site:

Neither of these statements make sense in the context of statistical theory.

How so?

Because you don't win fantasy contests by racking up some arbitrary player rating points; you win by scoring the most daily fantasy points. So it doesn't really make sense to say

This makes it seem like there's some other thing that is not fragile and does have an approachable level of certainty to trust (whatever that means). But if such a thing existed, it would have to be highly-correlated to actual fantasy scores, or at least more correlated to the actual than whatever model(s) it is outperforming. If that's the case, then you could just add it to a model that predicts actual scores and see dramatic improvement in the estimates and prediction intervals.¹ If these ratings are any good at beating dfs, then they are basically just transformations of accurate point projections.

I don't understand the logic here. Why can't you backtest models of predicted fantasy points against actual ones? He says this is the second reason why they don't offer point projections, but there's absolutely no methodological issue in backtesting the actual fantasy point values.

¹Assuming the model is correctly specified.

There are over 20 public/pay models and the quality ranges immensely. Even in MLB some are good at pitching, other hitting, none are great at both (well maybe one.)

Lol @ calling FantasyLabs donks.

You can use their models to back fill against projected and actual points. Hell, that **** is free using the Trends tool on their site.