Intro to the MLB DW Player Prop Sheet
I show and explain the in-season player prop betting analysis sheet
Probably the two most interesting and valuable things I do here are
The Daily Notes
The Daily Projections
Let me give you a quick overview of the projection model to get us started.
Projection Model Overview
I have developed and have been tweaking a daily MLB projection model. It was designed with DFS in mind, but it’s useful all over the place. To get to the final fantasy point projection, I project every category along the way.
It starts with player skill projections. These project what the player would do in a number of categories in a neutral environment. My script runs twice per week and updates baseline skill projections for every hitter on a 40-man roster. I do that for strikeout rate, walk rate, stolen base attempt rate, stolen base success rate, the full batted ball profile, and then it has some splits adjustments for each guy as well.
Every day during the season, the daily projection model takes those skills and layers on top of them that day’s environment. It looks at:
Lineup spot & context
Ballpark
Opposing starting pitcher
Opposing available bullpen
Weather*
One thing that I don’t have (yet?) is the daily weather forecast. It’s currently working just by putting a general weather factor in based on what time of year it is. Hitters project better in August than in early April, for example. But it does not account for wind and stuff, which honestly I think is fine because I feel like that business leads you down the wrong path more often than not.
It’s a good model. I’m not going to come out and say that I think it’s better than The Bat, because I’m guessing it’s not. If you’re really looking for the best available daily MLB projections, I refer you to that link. You’ll have to pay a lot more for those projections than for mine, so I do think mine might be the better value.
Enter The Player Prop
Anyways, one way we can use these daily projections is to exploit player prop betting. There’s no profitability in betting MLB totals or game lines. The lines are too sharp, and the juice/vig/rake/whateveryouwannacallit takes away any possibility for being +EV.
What some people are able to do is become profitable in the lesser known markets. The individual player props.
How It Works
So I grab all of those lines (from multiple sports books) and then compare my model’s projection with them to find the best bet values of the day. I’ll tell you a bit about that here.
The way it’s displayed is what I call a bet value. Here’s an example from the postseason last year. We’re looking at starting pitcher strikeout lines:
Let’s walk through that column by column, looking at the first row.
Player: Jack Flaherty is the pitcher
Game: The game is Dodgers at Mets
Book: This specific bet line is on FanDuel
O/U: The specific bet was to take the under
Line: The strikeout line was set at 6.5
Price: If you took the under 6.5, you’d get -142 on that bet
Projection: My model’s projection was 5.8 strikeouts for this game
LineOdds: The -142 price implied that there was a 58.7% of chance of Flaherty being under 6.5 strikeouts this game
ProjOdds: My projection suggested that the true odds of Flaherty being under was 63.8%
BetValue: This is ProjOdds minus LineOdds, in this case a 5.1 point difference, so a +5.1 bet value
This is the proper way to do this mathematically. A lot of people out there on Twitter or wherever else trying to sell you their picks will just telling you that a guy is in a smash spot and therefore that you should bet the over, with no regard to the betting odds or the actual chances of things happening. Having a decent projection model is necessary to have any chance of being a profitable bettor. This is not to say that you will be a profitable bettor with my projections, but at least it’s possible.
Pitfalls
The problem is, of course, that sportsbooks are smart, and they rig the game so you lose. Their lines get sharper every year. We saw an example of this last year. We were crushing them on hitter strikeout props. Their lines were just improperly done, and every day, my model was spitting out a handful or two of profitable unders. It would never recommend an over. So we were betting unders every day and profiting. But they realized their error in the middle of the season (multiple books did this) and took down the props entirely. They showed up again a few weeks later, and my model then stopped recommending so many unders because they had strengthened up their lines.
That will continue to happen. And the real backbreaker is that if you’re actually taking a significant amount of money from them, they’ll just limit your play or outright ban you.
There are professional bettors out there, so it’s possible, but it is not a life I’d recommend for anybody. You’re playing against the system, and the system is really hard to beat. If you’re looking to make money, there are better ways to do it.
So that’s not the approach I take or that I recommend. We’re just here to have some fun watching baseball, and using the algorithms and data models to make a little bit of cash along the way.
Bet Types My Model Calculates
The model does exactly what I just detailed above for these bet types (and I’ll add more if more are made available):
Pitcher strikeouts
Pitcher walks allowed
Pitcher hits allowed
Pitcher outs recorded
Hitter strikeouts
Hitter walks
Hitter singles
Hitter doubles
Hitter home runs
Hitter steals
Some of those work much better than others. Hitter home run props are the most popular, so they’re juiced beyond reason, and that makes it pretty foolish to bet on.
Every time the projections run (usually every 20 minutes during the day), it refreshes this player prop betting sheet and then piles up the best overall bet values in the “All Bet Values” tab.
I also have a “Recommended Bets” tab. The rules on that change throughout the year. I will work to keep the filters on it to only show the bet types that have been profitable so far in that season. For example, my model was poor at pitcher outs props. The math does not quite work the same when it’s a category where you can gain multiple of the thing in a single play (a double play). The explanation of that is complicated, and I don’t fully understand it myself. But just know that I’m trying to figure out what’s working and what’s not working during the season and then filtering down the recommended bet types based on that.
The many users of it who have told me they’ve had success with it have found their own ways to use it.
What To Expect
This sheet will be populated from Opening Day on throughout the season and available to paid subscribers. There will also (very likely) be prop recommendation articles on this Substack that hit your inbox at some point in the afternoon. That won’t happen every day, but it is something I’m trying to do more of this year. I’m probably even going to be bringing in someone else here to help me out with it.
I also track the bets as we go. Every bet is logged and compared with reality the next day, and I save it all off for analysis so we can see what’s working and what’s not.
Again, I’m not guaranteeing anybody success here. But if you’re going to bet on baseball, you should do it in a smart way, and this is one way to do that.
The sheet will be totally free for all to check out during the season’s first weekend before going behind the paywall. So check it here in a couple of weeks when we get started! I wanted to write this to answer the many questions I get asked about it and to set the stage for another very fun 2025 season.