Hey all, I’m wondering if you guys have had more success betting tournament matchups vs round matchups? I know round matchups are much higher variance, and in a small sample size I’ve done very poorly in them. Usually I try to get a line $0.20 or better for tournament matchups and $0.30 or better for round matchups when using the datagolf model. Should I lower this range on both to lets say $0.10-$0.20? Curious to know what everyone else is doing in determining which matchups to bet. Thanks all and good luck betting!
Datagolf did a study on this and they recommended a EV threshold of +0.08 for best results.
Since then they may have performed another (unpublished) study lowering this EV threshold to +0.07.
I’ll let the owners of the site chime in on this one.
For the win and place markets I’ve used a threshold of +0.08 for odds of less than for +1000. I raise that to +0.10 to +0.20 for odds of +1000 or more, using my discretion as to whether to take the wager or not. I always take the wager if it’s +0.20 or better.
1 Like
In practice the actual edge is about 1/3 of the expected edge on average and the +0.20 seems to do no better than the +0.10. Hence this throws the Kelly calculations out the window too.
Can you elaborate on this more? What exactly are Kelly calculations?
Well this is the Cliff’s Notes version of Kelly:
- The stronger the edge, the more you bet
- Bet bigger on favorites, bet less on underdogs
- Betting the right bet size maximizes the speed of bankroll growth. Any other size will result in losses or slower growth
1 Like
Here is the intermediate version:
- Kelly wager size can only be calculated definitively if you know what your edge is. If you don’t then you can easily overplay your edge
- If you have a +EV bet, underbetting will lead to slower rate of bankroll growth, but will always lead to growth. Overbetting can lead to losses, even with nothing but +EV bets
- In general you want to bet way less than what the Kelly % says. In practice I’ve found that you do not want to bet more than 1% of bankroll on any single wager (Unless you’re betting big chalk)
- Also if you bet big chalk then you would be lowering the +EV threshold. For example +0.02 is plenty for a -500 wager and +0.01 is excellent for a -1000 wager. For a -200 wager I’d be looking for something like +0.05
1 Like
And here are the recommended wager sizes using Kelly Principles:
Wager size = 1%/(Decimal odds - 1)
So an even-money wager would have odds of 2.00
Wager size = 1%/(2.00 - 1) = 1%
Let’s say you’re betting on Scottie Scheffler to win at +1000, or 11.00
Wager size = 1%/(11.00 - 1) = 0.1%
You wager on Rory to make the cut at -400, or 1.25
Wager size = 1%/(1.25 - 1) = 4%
You bet on Charles Howell III to win at +10000, or 101.00
Wager size = 1%/(101.00 - 1) = 0.01%
1 Like
Thanks Daniel! This is good stuff! Do you personally have better results in tournament H2H’s compared to round H2H’s?
I rarely play tournament H2H and round H2H are not common either.
I have better luck finding value in the 3 balls.
Tournament 3 balls are pretty good when you can find them.
In general you want props with 3-10 players, that’s where Datagolf seems to really excel. The place markets is another gold mine
Oh I see. I would think that tournament matchups would very lucrative. The place market being a “gold mine” is interesting. Are you referring to mainly T5, T10, and T20’s? Right now when I sort by EV on DraftKings for T20 there are 9 golfers with an EV of 0.20 or more. Does this mean you are betting all 9 of them? Similarly to outrights, on Fanduel there are 7 golfers with an EV of 0.20 or more to win outright. Are you betting all of those as well?
I’m primarily a H2H bettor and I’ve been successful. I’m going to dabble in the place market as well, so I’m trying to get a feel for it. Do you kind of put together a portfolio of bets each week with various outrights, t20’s, and 3 balls? Again, thanks for your time and the info.
1 Like
I find that tournament matchups are hit-and-miss, I’m lucky to get 5 a week and my EV is probably like <5% on those. I prefer tournament 3-balls and groups where I have better opportunities to get good prices.
Similar story with round matchups, hard to get value on those. 3 balls are definitely better. The prop bets (things like low Australian, etc.) are very good as well and the place markets (T5/T10/T20/T30/T40) is where the majority of the volume comes from.
Cut market is another avenue but I find that DG is very inconsistent on those. I think it’s losing money overall but I’m not sure.
You also really need to pay attention to the adjusted win and place odds. There’s a real gold mine after Round 1 and to a lesser extent after Round 2. The books add a ton of juice after Round 3 so your opportunities are probably gone by then.
I am betting all the place and win odds of EV 0.20 or more, and many with smaller EV’s (depending on the odds I’m laying). Yes, I’m betting them all. This is another reason why I bet like 10-15% Kelly. If Oliver Bekker is highlighted by DG I’m going to be on him on the win market, place markets, matchups, 3-balls, Groups, Low South African, etc. I need to avoid overexposure since it’s possible to have like 5-10 bets on the same golfer.
I’ve always thought the cut bets were the among the most profitable, at least at Bet365. Books consistently don’t adjust their cut offerings when their is a weird distribution of skill (e.g. very low-quality bottom 25% like at the Puerto Rico Open resulted in 10-20 bets on the top guys to make the cut).
Bet365 may be a special case, the other books usually only offer the top golfers and for the most part the lines are tight
I almost always end up betting them to miss the cut and I’m left disappointed fairly often
I guess that’s why they are top golfers, they always seem to find a couple of birdies on the back 9 in R2 to sneak into the weekend
Great stuff. So for the prop bets you are just simply checking the odds on DG and then comparing them to other sites and betting them based on how far off the EV is? And then for group betting you are using the group simulator on DG and then doing the same thing with EV?
Yes sir
It takes a while to get like 200 bets in, LOL
It’s a race to get as many bets in as possible before the inevitable line movement
3 Likes
Curious if the EV >= .07 is in fact optimal. I would expect the threshold to change based on the type of bet and the likelihood
As far as the sizing, from whatever your minimum EV threshold is you can scale up from there, but you’ll need to put a cap on at some point. I personally use .3 in a kelly calculator with adjustments for correlated bets Kelly Criterion Bet Calculator: Optimize Sports Betting Bet Size My cap is 5% for other sports but I haven’t quite worked out what it should be for golf yet
I also think it’s important to not bet the recommended amounts blindly. Betting a lot on a longshot or on something with a low return usually isn’t the best idea, and there’s a thread in here somewhere on not doubling/tripling down between rounds on the same bets already placed earlier
Are you primarily betting Round 3 balls? I don’t see anything on DG about tournament 3 balls. Thanks again!
I’m betting whatever I can find using the simulator.
Round 3 balls are great but the line movements kill me. Much better chance of being able to grab Tournament 3 balls.
Some markets are more sensitive to line movement than others.
1 Like
By the way I did provide a Kelly-inspired formula for bet sizing on an earlier post above and you can use that formula blindly and do pretty well.
At some point though you’ll get limited by the books if you’re having success so you’ll just be pushing those max wagers, LOL
“Death by a thousand cuts” is my motto and I try to get in about 1,000 golf bets a month. If I have an average bet size of $20 and an ROI of 5% then you’re looking at $1K/month.
Yes, I think I’m going to start using that. I jumped on a ton of the DK top 20 odds on those guys. Correct me if I’m wrong but for a $50,000 bankroll I should be risking $500 for bets around EV. Then lets say I bet on Hickok +1200 to Top 20. On that bet I should only be betting $42? Essentially to win $500 on all those T20 bets?