Building a $249 NFBC Offense

As we continue to maintain hope that there will be a baseball season this year, last week I decided to have a bit of fun and build a $9 pitching staff using NFBC average auction values calculated from March 15. While that was a joy, commenter runningfrog demanded that I share what a $251 offense would look like. If I were to save all those auction dollars by buying such cheap pitching, how incredible might my offense look?! I decided to find out.

I began by opening my LABR mixed leagues values that I had calculated in early February when the snake draft was held. Right now, it’s the only mixed league values I have put together. The good thing about my values is that the calculations are tied to my Pod Pojections, so the values are updated with my most recent forecast set. I then compared my values to the NFBC average auction values. My strategy was generally to roster the hitters I was projecting to earn the most profit (Pod Projected value versus NFBC average auction value). Of course, I couldn’t solely do that, as I wanted to spend the full $251 and also needed to fill the 14 positions.

I didn’t pay too much attention to categories, assuming that I’ll end up buying a mix if I was just buying the biggest profits. However, I did shy away from the hitters whose values are driven more by a strong batting average than their counting stats.

Funnily enough, even knowing how much players cost, I still ended up leaving $2 on the table. After finishing my team, I went back and looked at every upgrade opportunity. In every possible case, my only options were players I valued less than the player I originally chose to roster, yet they cost more.

As a reminder, I’m displaying ATC projections, but actually used my Pod Projections and values to assemble the team. As you could guess, I’m more bullish on the majority of these batters than ATC is. Obviously, these projections are based on a full season of play.

So, please enjoy this sweet, sweet $249 offense.

The $249 Offense
Pos Player AVG* HR* RBI* R* SB* AAV
C Christian Vazquez 0.261 14 53 52 4 $5
C Jorge Alfaro 0.250 16 52 43 3 $5
2B Jose Altuve 0.303 24 82 96 11 $25
SS Adalberto Mondesi 0.251 15 66 70 46 $29
MI Whit Merrifield 0.286 14 63 85 22 $21
1B Freddie Freeman 0.298 34 105 100 6 $31
3B Nolan Arenado 0.298 38 114 99 2 $35
CI Josh Donaldson 0.261 33 93 89 4 $16
OF Charlie Blackmon 0.299 28 79 103 4 $23
OF George Springer 0.276 34 90 107 7 $22
OF Joey Gallo 0.233 41 91 86 6 $15
OF David Dahl 0.281 22 70 71 7 $11
OF Mike Yastrzemski 0.247 22 64 75 4 $1
Util Shohei Ohtani 0.279 22 70 61 11 $10
Totals 0.275 357 1092 1137 137 $249

Surprised? Disappointed? Simply looking at the names, it certainly doesn’t seem like a loaded team. I’m also not even sure if these projected stats would figure to finish first in any categories. ATC and Pod Projections were identical for steals, while ATC was slightly lower on batting average and significantly lower in home runs, RBI, and runs scored.

Buying catchers was interesting. I know that everyone values catchers differently and the vast majority of the time, a catcher is bought for less than his calculated value. Amazingly, out of all catchers bought for at least $2, just one of them was overvalued. This was Roberto Perez, who only went for $2, and I had him valued at $0.49! Literally every other catcher was underpriced compared to my values. With Christian Vazquez and Jorge Alfaro, I bought the two with the highest dollar profit, and only needed to spend $10 combined. I could have paid $22 for J.T. Realmuto and netted some profit, but opted for better stats from similarly priced players at other positions.

Middle infielders were almost universally overvalued. It’s really bizarre because it almost seems as if auctioners are double counting position and giving them far too great of a boost. Of the top 10 middle infielders, nine of them were bought for well above my value. The lowest overvaluation was about $4 (Javier Baez), while the highest was a whopping $14 for Keston Hiura (sold for $26!). Adalberto Mondesi and Jose Altuve were the only two MIs that sold for at least $25 that I calculated as going for exactly fair value. That is precisely why they joined the Pod Squad.

Neither of these two surprise me as being fairly valued overall and a much better deal than the rest of the top tier. I actually drafted Mondesi in LABR and I’m more bullish on his power than everyone else, though we won’t know how his shoulder surgery might affect it. The season delay will give him more time to get to full health. Between Altuve’s loss of stolen bases and possibly the Astros cheating discount, he comes at a reasonable price for the first time in years. He’ll still only be 30 this season and you have to imagine that better health should result in some sort of steals rebound, at least back into the teens. I find Whit Merrifield annually undervalued, and I’m not entirely sure why. He’s a near five-category contributor, with RBI being around neutral given the difficulty of amassing them as the leadoff hitter.

I had to spend my money somewhere, and it went mostly to top tier corner men. Oddly, unlike the middle infielders, corner guys were mostly fairly valued or slightly undervalued, which made them a good position to spend the going rate. I found that both Nolan Arenado and Freddie Freeman were bought at a slight discount, which doesn’t often happen for top two round hitters. Both are elite four-category contributors. According to my values, Josh Donaldson at $16 was the best value among hitters that were sold for at least $10. While I worry about age-based regression, I think he’s in a great situation (more favorable home park and excellent lineup) to offset any such effects.

Hmmm, I definitely went into this exercise expecting to buy a more exciting outfield! Charlie Blackmon stopped stealing bases last season, but remained an elite four-category contributor. Any sort of steals rebound will lead to ample profit here. George Springer is one of the few hitters that my projections are actually slightly more bearish on than ATC. So there’s actually even more profit potential here than I calculated if he comes closer to ATC than mine. While Joey Gallo’s batting average downside is scary, one can only dream of how many homers he would hit in a full healthy season. Do you realize that his career high ABs in a season is just 500? Obviously, he won’t set a new record this year, but the home run pace is all that matters when/if the season gets going.

It’s difficult to project David Dahl because he’s always injured, but he’s a worthy gamble at just $11. I’m bullish on his power, he’ll steal a handful of bases, and he owns legit high-BABIP skills to support a strong average. I hate to label anyone “my guy” or something, because really, I usually can only tell you after a draft if someone is “my guy” since it’s solely based on how much more I value a player than that specific league. Before my draft or auction, I cannot possibly know who those players are! And yet, it seems pretty obvious that Mike Yastrzemski is one of “my guys”, but only because my projections are much better than the rest of the systems and he went for just a buck in NFBC. He should play every day in that weak Giants outfield, perhaps hit leadoff, and have every chance to prove his power spike was real.

While I’ll never understand how a person could be so talented at baseball, Shohei Ohtani is, whether it makes sense or not. We aren’t really sure about his pitching this season, but he has proven he could hit with the best of them. Sure, he’ll clog my Util spot, but I calculated some serious profit at just $10. In fact, he was the second most profitable hitter in my calculations.





Mike Podhorzer is the 2015 Fantasy Sports Writers Association Baseball Writer of the Year. He produces player projections using his own forecasting system and is the author of the eBook Projecting X 2.0: How to Forecast Baseball Player Performance, which teaches you how to project players yourself. His projections helped him win the inaugural 2013 Tout Wars mixed draft league. Follow Mike on Twitter @MikePodhorzer and contact him via email.

13 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Werthlessmember
3 years ago

“I’m also not even sure if these projected stats would figure to finish first in any categories. ”

What do you suppose causes this, and isn’t this a problem?

Personally, when I see this team, my suspicion is that it doesn’t “feel” elite because you feasted on the mid-value guys. Look at how many players you got in the low twenties. If I were attempting this strategy in real life, I would have more $1 players. I’m not saying the strategy for filling the roster did not maximize profit, but it might not have… because you forced yourself to limit the $1 assets. If you used a modified stars and scrubs, could you get better overall stats, even though you “earn no profit” on Trout/Acuna/Yelich?

Rotoholicmember
3 years ago
Reply to  Mike Podhorzer

Agreed. Stars and scrubs is good for leagues where you expect to churn the last few roster spots. But in an exercise like this its purely about value. There’d be no way to significantly improve the projected stats there while spending the same amount of money. Could maybe eke out a few extra bucks by optimizing a bit more, either using an algorithm or the Excel solver function which I’ve come to love. But it wouldn’t make much difference, and I doubt if an optimized lineup would use more stars rather than less.

Jeff Zimmermanmember
3 years ago
Reply to  Mike Podhorzer

You don’t keep your LABR draft standings?

couthcommander
3 years ago
Reply to  Mike Podhorzer

Because you’re using average dollar price, there’s a super long tail of cheap players ($4 or less) that would almost all cost $1 in a live draft. I think that’s one possible explanation of how your optimized team can’t really match a stars and scrubs approach. If you assume some extra value (say two players you value at $6 that average $4 but might only cost $1), you’d have an extra $10 to invest in a player(s) with worse profit but better stats.

Werthlessmember
3 years ago
Reply to  Werthless

If you were trying to maximize profit, your roster would have a lot of $1 players… based on distributions posted in other articles, the expected value of some $1 players is $5. That not only gives you a great margin, but gives you a good profit per player. The problem is that if you filled out your roster this way, you’d use a fraction of your budget, and your overall expected value would be too low. You’d spend less than $100, and might get $200 in value. This is why I was suggesting that your manual approach, of focusing on $10 players, might not be producing the stacked roster that you would expect… you should seek out the low value contributors. To fill out out a lineup that competes in all categories, you may need to pay sticker price on players who also contribute steals (eg. Acuna). You may not earn the highest profit, but your expected roto rankings might be higher.

As another commentator said, if you have a dataset with expected profits and cost per player, you could either solve for it, or run a simulation with certain constraints, to figure it out. Even more interesting, perhaps, if you were maximizing roto points using data on average standings values.

Edit: For example, Arenado+Blackmon might underperform Cron+Acuna, even if Acuna expects to produce the lowest profit of the 4.