Draft Day Talent: 2017 Tout Wars Example

It’s tough to create a perfectly balanced team on draft or auction day. Owners are feeling the push-and-pull of trying to balance all five categories in a roto league. Mid-draft, many owners decide to drop a category with the hope of finding the needed stats on the waiver wire. Knowing which stats can be found can be tricky. By looking back at last season’s Tout Wars leagues, a decent idea of available stats can be determined.

One feature of the OnRoto.com fantasy league website computes the league’s final standings using just the drafted teams (nine pitchers, 14 position players). I took these draft values and compared them to the actual final values for each of the four roto leagues (12-team AL and NL-only and the two 15-team mixed leagues).

Some specific notes on these leagues. First, they are deeper than most leagues so every player who might be good is already owned. As for the timing of the mixed draft (the other three were auctions) happened a few weeks before the other three. Additionally, only the 23-man rosters were used used for the projected standings. Each team had an additional five or six-person bench.

With the technicalities out of the way, here are the percentage change in counting stats from the final standings and the initial rosters. I will look at the rate stats later.

Notes:

• In these deeper leagues, around 75% to 90% of the production is from the draft rosters. With hitting, all but 15% of the stats came from the starting rosters. Assuming each of the 14 hitters produced the same (bad assumption but follow along), that means two of the 14 hitters (2/14 = 14.3%) were replaced. Likely the hitters were lower quality so the number may be one or two higher.
• I find it interesting that 89% of the stolen bases accumulated came from the initial rosters while the home run numbers were down at 85%. At least last year, it was easier to find home runs compared to steals.
• The mixed draft’s percentages were a bit lower than the mixed auction (besides Saves). This is not surprising with the owners in the auction having more information available on playing time and injuries.
• Of the pitching categories, Saves are surprisingly the most unavailable pitching category with 96% of the NL-only Saves being on the initial rosters. I’m a little surprised by this number. It was almost as easy to find Home Runs as Saves across all the leagues. Assuming each team has 6 starters, they are all league average, and provide most of the strikeout numbers (again bear with me), owners hit on 4.5 (.77 *6) starters on their initial rosters. If assuming seven starters, they hit on 5.5. These are within reason.
• The Wins and Strikeouts final values are almost identical for each league but are the lowest rates just under 80%. This doesn’t surprise me either with many teams stocking up on starters to use in their two-start weeks.

Now for the rate stats. Here are the average final values along with the values from the initial rosters.

Notes:

• The initial roster and final on-base perpecentages were almost even.
• The initial roster WHIP and ERA were all lower than the final values. This points to teams streaming in subpar pitchers for the strikeouts and Wins. I went through the reserve round picks and found almost a perfect 50/50 split between hitters and pitchers. I thought the pitcher percentage may have been a little higher with the amount of streaming possibly going on.
• I find the tradeoff between WHIP/ERA and Wins/Ks interesting. Teams who kept their initial rosters would have seen their lower ERA and WHIP values gain them about 10 total spots on the rankings (~5 spots each). Were the extra strikeouts and Wins worth it? It’s impossible to tell. I believe it comes down to a tough balancing act in which excellent middle relievers are becoming more and more valuable by giving owners good rates and strikeouts. Wins are taking a back seat.

Conclusion

It’s tough to tell which stats can be found off the waiver wire in these leagues. For hitters, it seems to be tougher to pick up Stolen Bases compared to Home Runs. For pitchers, it’s tough to tell. Saves may be the hardest to find with starting pitcher streaming messing up all the other stats. If I was to intentionally not spend resources on an area, it would be for low end starters. Most of the other stats are being found at similar rates from pickups.