Over the course of the last several weeks, I’ve turned my sights more to preseason drafts. In recent articles, I’ve mostly been evaluating players based on how interested I would be in drafting them at their draft cost (a price which I can only assume at this point).
It’s possible, even likely, that as many of you read my player assessments, you’ve had a moment (or two, or a thousand) when you’ve said, “What is this guy thinking?” Fair question, so all season I’ve tried to highlight some of the data that I think is most helpful in making calls on which players I roster, pick up on waivers, and drop. While I consider other factors that I don’t spend time writing about, the underlying stats I write about each week are probably the main assessment tools I use during the season.
In points leagues, it’s a relatively easy process to project player stats and then apply an estimated point value that player seems likely to earn for the season. Today, I’ll focus on Roto since it involves a few more steps, in my opinion.
In Roto, those underlying stats are extremely helpful when I’m in the middle of a season, and I know which categories I need to bolster in order to get the best overall stats across all 10 categories. But what about during draft season? There are no standings then, so we don’t have the ability to look at the SB category and see that we’re only 4 SB away from picking up another point in the standings. In season, when we see a category where it’s possible, even likely to gain points, it’s an easy decision to go get someone from the waiver wire who can get us there without hurting us in other categories. But in the preseason, we are deciding between any number of hypothetical rosters. How do we decide between two players of similar projected $ value?
This week, I’ll go through at least some of my pre-draft process to determine why I usually prefer certain types of players more than I’m concerned with overall projected $$ value. To give a realistic example of what my decision-making looks like, I’m going to review 3 players that I mostly faded during last year’s draft season and will likely fade again this year:
Xavier Edwards, 2B/SS, MIA
Brandon Lowe, 2B, TB
Bryson Stott, 2B, PHI
Since going through my entire process would bore most folks to tears, I’ll condense it into just a few main steps:
1. My initial expectations of a player
I always want to have my own understanding of a player’s performance level so that I’m not just blindly trusting other people’s projections. Don’t get me wrong: projections are great. They are a major part of my process, and the tools and formulas those systems use are much better than I am at predicting overall values across all likely players in any given season. But if I don’t have any understanding of a player’s skills, then I’m never in a position to find potential value in those cases when systems rate a player lower than how I value his skillset. That can and does happen, and I want to take advantage of the opportunity when it’s presented to me.
2. What projections say
But I’ll definitely have lots of examples of projection systems that I reference, compare, average, you name it. There are those I like better for hitting and ones I prefer for pitching, so I may give certain systems priority at different times. But overall, I try to take a reasonably wide view of 5+ systems. Maybe that’s too many, but it allows me to see where the systems consistently agree with each other AND where they disagree about any given player. I take an average of the 5 or 6 systems I like the best, but pay close attention to any spots where 1 or 2 systems are vastly different from the overall average of projected values.
For today’s exercise, I’ll use 2 of the projections I like: Razzball and BatX.
I also look closely at projected auction values of those systems. But it’s important that I keep an essential factor in mind: Different systems assign different $ values to particular skills, so it’s possible that some of the predicted values show surprising variance when compared to each other. That’s another reason it’s important that I have my own sense of what I can expect from the player – when projected $ values for a player show significant variance, I have a better idea of which one makes more sense in the context of my own expectations.
For today’s exercise, I’ll use these two systems for auction value: Razzball Preseason Player Rater and Fangraphs Player Rater released during the preseason.
In an attempt to end my need for other projection systems’ auction values, I’ve tried different formulas to help me determine my own $ value. I’m not exactly a mathematician, so those attempts have been met mostly with extraordinary levels of failure. But if you’d like a good, detailed description of how to create your own $ values (that even I can understand), I recommend Larry Schechter’s book Winning Fantasy Baseball.
3. How those projected stats would likely impact my team
There are many ways to try to assess how a player’s skills will affect your overall stat line, and I encourage you to do some research in Standings Gain Points and Percentage Value Method, among others.
For the sake of this exercise, I’m going to consider what it would look like to reach the 80th percentile in each of my 10 stat categories. Theoretically, if I’m in the top 20% of a league, I’m probably going to be competing for one of the top 2-3 spots. To do this, I need to know the league history of the league(s) I’m playing in, so it’s key to research the average season results in your league, particularly in the last couple of years.
I tend to play NFBC leagues, so I’ll use information published prior to the 2025 season from The Process, an incredibly helpful yearly publication put out by Jeff Zimmerman and Tanner Bell. According to their research, the top 3 spots in a 12-team Online Championship (which would mean 75th percentile and above) are
| Year & Finish | AB | AVG | HR | R | RBI | SB |
| 2024, 1st | 7,568 | .2628 | 335 | 1,139 | 1,114 | 237 |
| 2023, 1st | 7,524 | .2686 | 346 | 1,182 | 1,150 | 233 |
| 2024, 2nd | 7,548 | .2592 | 318 | 1,105 | 1,076 | 214 |
| 2023, 2nd | 7,481 | .2652 | 331 | 1,147 | 1,118 | 211 |
| 2024, 3rd | 7,474 | .2572 | 308 | 1,085 | 1,052 | 202 |
| 2023, 3rd | 7,459 | .2629 | 322 | 1,125 | 1,098 | 197 |
If I want to be at the 80th percentile, it looks like I would need roughly
| AB | AVG | HR | R | RBI | SB | |
| 80th% | 7,475 | .260 | 315 | 1,110 | 1,100 | 205 |
| Per player
(14 total) |
533.9 | .260 | 22.5 | 79.3 | 78.6 | 14.6 |
I suspect those three steps sound kind of obvious – I admit I don’t have a secret formula or the kind of ability to scout players’ potential that will lead me to league win after league win. My approach is pretty standard, but it works for me (and lots of other players), so maybe it’s worth a closer look.
Xavier Edwards
With Edwards, my initial assessment is that he’s a possible plus in AVG, an absolute albatross in HR, a minus but potentially unharmful contributor in R, a minus in RBI, and an absolute plus in SB. If you’re familiar with Edwards’s game, you might think I’m not bullish enough on his AVG potential, but his lack of hard-hit ability makes me worry that his good performances in AVG are mostly probable but not a lock. If he consistently puts up HH percentages in the 20s, that makes me think that the BABIP gods could turn against him at any moment.
When I consider Edwards’s 2025 projections:
| AB | AVG | HR | R | RBI | SB | $$ | |
| Razzball | 506 | .279 | 5 | 68 | 43 | 32 | $10.20 |
| BatX | 515 | .276 | 2 | 67 | 40 | 31 | $1 |
Edwards is a great example of a player who can be seen completely differently depending on the types of stats a manager looking for. The surface stats for Edwards are practically identical, yet Razzball put him at $10.20 while BatX put him at $1. Wow. My initial feeling toward Edwards makes me lean toward the BatX HR projection, so I’ll use their HR number, but the Razzball stats for the other stats to create the projection I’m working with:
| AB | AVG | HR | R | RBI | SB | |
| Edwards | 506 | .279 (141/506) | 2 | 68 | 43 | 32 |
| 80th% goal | 7,475 AB | .260 | 315 | 1,110 | 1,100 | 205 |
| Per player
(14 total) |
533.9 | .260 | 22.5 | 79.3 | 78.6 | 14.6 |
| New Avg per player (13 players) | 536.1 | .2589 | 24.1 | 80.2 | 81.3 | 13.3 |
Look at the damage Edwards would do to the needed totals if he performs as projected. He helps AVG a bit and reduces the SB load even more. Every other needed stat is now significantly worse for me. As a result, he’s just not my kind of guy unless I’m desperate for steals. This was especially the case considering I was watching Edwards get drafted in the top 150 fairly regularly, and there’s just no way I want to intentionally drop so many of my categories at such an early point in the draft.
Brandon Lowe
I’m a Lowe fan. His power is real, and with that power, he will probably provide a decent number of RBIs. Probably. On the downside, he is likely an albatross to AVG. And since he is the strong side of a platoon, his playing time isn’t as great as Edwards’s is. Oh yeah, Lowe misses time every season for injuries.
When I consider Lowe’s 2025 projections:
| AB | AVG | HR | R | RBI | SB | $$ | |
| Razzball | 459 | .234 | 24 | 68 | 67 | 6 | $11.10 |
| BatX | 420 | .239 | 21 | 59 | 64 | 4 | ($2.80) |
Again, the projected stat lines are relatively similar, but the Razzball system weights stats in such a way that Lowe is expected to be even more valuable than Edwards while BatX expects Lowe to be of negative value. I definitely lean more toward the Razzball projected $ value, especially at the draft price I would expect him to have (post-200).
Since I tend to think Lowe will miss time, I’ll use BatX’s line as my baseline to work from since it gives him a lower total of AB. I’ve bumped Lowe’s AVG down a point so that it mathematically works (100/420 = .238).
| AB | AVG | HR | R | RBI | SB | |
| Lowe | 420 | .238
(100/420) |
21 | 59 | 64 | 4 |
| 80th% goal | 7,475 AB | .260
(1945/7475) |
315 | 1,110 | 1,100 | 205 |
| Per player
(14 total) |
533.9 | .260 | 22.5 | 79.3 | 78.6 | 14.6 |
| New Avg per player (13 players) | 543 | .262. | 22.6 | 80.8 | 79.7 | 15.5 |
Lowe technically would hurt me everywhere, but I would be getting him for much cheaper. As a result, I’d be interested in drafting him at the right price. I would also favor him because I think HR are harder to find off the waiver wire than SB. If I can get Lowe somewhere after a 250 ADP, I’ll probably draft him.
Bryson Stott
Stott is interesting since he plays for a really strong team. I find that I’m willing to draft him, but I’m also concerned that he is going to cost more than I want to pay for how much he might damage my stat line. He is also a strong side of a platoon, so that could hurt his AB total. Still, I would expect him to have a R total that would end up pretty close to my needed average. Otherwise, I think he’s probably a slight minus in AVG, a full minus in both HR and RBI, and a plus in SB.
Let’s see what the projections say:
| AB | AVG | HR | R | RBI | SB | $$ | |
| Razzball | 489 | .253 | 12 | 60 | 57 | 24 | $8 |
| BatX | 510 | .256 | 11 | 70 | 54 | 24 | $3.10 |
Both projection systems like Stott’s AVG better than I expected, so it’s good to have that push back to what I originally thought. Otherwise, the numbers are what I would expect. Since I think Stott will have more than 500 AB, I’ll go with BatX again. Once again, I dropped the AVG slightly to make the math work.
| AB | AVG | HR | R | RBI | SB | |
| Stott | 510 | .254
(130/510) |
11 | 70 | 54 | 24 |
| 80th% goal | 7,475 AB | .260
(1945/7475) |
315 | 1,110 | 1,100 | 205 |
| Per player
(14 total) |
533.9 | .260 | 22.5 | 79.3 | 78.6 | 14.6 |
| New Avg per player (13 players) | 535.8 | .261 | 23.4 | 80 | 80.5 | 13.9 |
As it turns out, the projections tell me that Stott is going to be less damaging to my numbers than someone like Edwards, and I can probably draft Stott cheaper than Edwards. Hmm, maybe I should have drafted more of him this season based on those numbers.
Keep in mind that, for this exercise, I’m not considering these players’ actual performance this year. So, if I were about to go into the 2025 season, this would have been one part of my preseason process. I also would need to consider what stats I’m willing to pay for at certain draft prices. As it turned out, I wasn’t super excited about these players at their costs because their stats are unbalanced (high in 1 or 2 areas, low in others). As I look at these numbers again, I should have been more willing to draft Stott, but I’m glad I faded Edwards.
I hope you found this helpful. I did at least.
Until next week. – ADHamley
