Who needs a time machine when you have the Projections and Pricing Tool?
So I’ve got some good news and some….well…it’s not bad news so let’s call it not-so-good news. The good news is I have access to some new splits data that will improve the Fantasy Alarm DFS Projection and Pricing tool. The not-so-good news is I’ve spent the past several days reworking the double-secret Excel code and the results of the recent research presented using the projection engine are slightly different.
Please don’t fret as the take-home lessons are all still applicable. The main alteration is the extent of the home versus away percent points change is a bit reduced. The reason for this is instead of applying a single home versus away coefficient, the engine now treats each stat independently and applies the necessary adjustment to each individual statistic.
By means of a quick review, last week the impact of a hitter being at home versus on the road and the influence of batter to pitching handedness were investigated. The study involved holding all the would-be variables constant except the one being tested. The associated projections were converted to fantasy points using the scoring of the most popular daily sites.
The primary advantage for the upgrade is not all of a hitter’s skills are affected equally be it home or away. So instead of applying a global adjustment, the benefits and penalties are now customized per player. The extent is minimal but it’s still best to work with the most accurate baseline as possible. Once the back-testing in complete (within a day or two) the new data will be presented.
But like Chip Diller says, all is well. In the meantime don’t worry. As has been alluded to, there is more to the Projection and Pricing Tool than simply sorting and using the players projected for the best bang-for-the-buck. The following is the first of what will likely be many tutorials on the most efficient use of the tool.
For the purpose of this discussion the focus will be on hitters. Pitchers deserve and will be afforded their own bandwidth.
Whenever one embarks on a discussion of this nature, a delineation between what’s best for tourney play and what’s optimal for cash games is usually necessary. This is no exception but for the sake of the K-I-S-S principle, we’ll talk in general terms and address using the tool for GPP and cash games will be forthcoming.
If you recall, the results of the points-projected study on Miguel Cabrera resulted in numbers in the low threes on FanDuel. Cabrera’s average points-per-game on FanDuel is currently 3.4.
We’ll use FanDuel since its scoring system is the simplest and easiest to think about without breaking out the spreadsheet.
- Each Total base: 1 point
- BB or HBP: 1 point
- Run scored and RBI: 1 point
- Stolen base: 2 points
- Each out: -.25 points
Here’s the bottom line. You aren’t looking for Miguel Cabrera to score 3.4 points; you’re looking for Miggy (and every player) to score considerably more than that. But 3.4 is his average which means for every huge game, there are a more below 3.4 to keep the average at that point. Think about it. A 2-for-4 night with a walk, double, HR, 2 runs scored and 4 RBI is worth 12.5 points. In the unlikely event Cabrera goes 0-for-5, he scores negative 1.25 points. The point being there’s a whole lot more points above 3.4 available than below so in order for the average to be 3.4, there has to be more games (a lot more) below 3.4 than above.
To wit, after 61 games, Cabrera has scored more than 3.4 points 24 times while failing to meet his average on 37 occasions. Here’s the actual breakdown of points per game (PPG) and the number of games that was scored:
| PPG | # games | PPG | # games | PPG | # games | ||
| -1 | 9 | 2.75 | 1 | 6.75 | 1 | ||
| 0.25 | 4 | 3 | 1 | 7.5 | 2 | ||
| 0.5 | 1 | 3.25 | 3 | 7.75 | 1 | ||
| 1 | 1 | 3.5 | 3 | 8.5 | 1 | ||
| 1.25 | 6 | 3.75 | 2 | 9.5 | 2 | ||
| 1.5 | 3 | 4.25 | 1 | 9.75 | 2 | ||
| 2 | 2 | 4.5 | 3 | 10.5 | 1 | ||
| 2.25 | 4 | 6.25 | 2 | 10.75 | 1 | ||
| 2.5 | 2 | 6.5 | 1 | 13.75 | 1 |
Granted, this is arbitrary, but if we set six points as our target for a good night from Cabrera, he only managed to get that 15 times. Only is said with air quotes since that’s a lot in DFS terms.
I don’t care who you are and how good your projection engine is. Unless the engine is in a Delorean that’s equipped with a flux capacitor and you have an endless supply of plutonium or can count on a lightning storm, you won’t be able to predict the games Miggy goes off. The best that can be done is to find opportunities for a big game.
THAT’s what the Fantasy Alarm Projection and Pricing tool does. It identifies the matchups that are more likely to result in a game much better than average. As such, the difference between 3.1 and 3.3 or 3.7 and 3.9 isn’t significant. You’re looking to be pointed in the right direction. The tool is your GPS.
For what it’s worth, the difference between the data presented for home and away and the adjusted data is in this 0.2 range which is why you shouldn’t be concerned about the change. When all the pieces are in place, that tiny amount won’t influence your decision. And if it does, you’re not approaching things properly.
OK, back to the purpose of the projection and pricing tool. While the ultimate goal is to beat the daily projection, it follows that a player has a better chance of having a great night when his projection for that game is above his average. So that’s the first utility of the tool. All the daily sites provide the average fantasy points per game for each player based on their scoring.
ONLY USE PLAYERS WITH A PROJECTED PPG GREATER THAN THEIR AVERAGE PPG.
The other primary use of the tool is highlighting the players with the best bang-for-the-buck. This is determined by taking their salary and dividing it by the projected points. The smaller the number, the better the projected bang-for-the-buck. This is a great way to uncover the value plays on each site.
WHEN LOOKING FOR VALUE PLAYS, USE THE PLAYERS WITH THE LOWEST $/PTS
WARNING: Occasionally the above bolded statements are in conflict. That is, if the site’s pricing is really, really low, there’s a chance the player has a below average game but still offers a favorable return on investment. In these rare instances:
ONLY USE PLAYERS WITH A PROJECTED PPG GREATER THAN THEIR AVERAGE PPG.
Why? Keep in mind the actual PPG projection isn’t as important as the relationship to the player’s average PPG. If on one hand the hypothesis is a projected PPG better than average is a harbinger for a good game then the opposite is true. A projected PPG worse than average can portend an even worse game.
Daily fantasy is all about upside potential. In fact, a Cliff Note’s motto for success is maximizing upside potential while avoiding downside risk. Hmm, sounds a lot like traditional fantasy.
The point is, even though a handful of players may look like they’re providing value, the downside risk isn’t worth it. Instead locate another option that offers both upside with a built-in bang-for-the-buck based on their price tag.
There’s a lot of stuff here to digest so let’s call it a day. In the future we’ll walk you through how to set a lineup using the Fantasy Alarm Projection and Pricing Tool.