Game Balanced Fantasy Baseball
If I have one complaint about fantasy baseball, it’s the importance of an early lead. Of course, it’s possible to scuffle through April and still win a roto league. In such a scenario, victory requires a heroic effort. The roto game is rigged to favor April’s top performers, and H2H isn’t much different. I’ll present an alternative today, but it does require some manual work from your commissioner.
First, why is the early season so important? In roto leagues, an early lead in the standings can open up various strategic and tactical opportunities. Fluky April power can allow you to convert J.D. Martinez into Dee Gordon without serious risk to your power related categories. Now you have more runs, stolen bases, and average.
An early lead can be especially valuable on the pitching side of the ledger. A lucky win total makes it easier to siphon more innings to relievers. As we’ve discussed recently, relievers can post anything from a zero to 35 win pace. Usage matters, but reliever wins are still mostly unpredictable.
An excellent performance in ERA and WHIP means you can manage your starters for wins rather than ratios. It’s not so hard to pile up the Ws when that’s the only category you’re maximizing. A huge early season lead in strikeout rate also gives you room to maneuver.
The H2H game is slightly different since your stats clear after each week. You won’t be able to exploit the same institutional advantages as a roto owner, but you do have one common factor in your favor – perceived value.
The first few weeks of the season can massively upgrade a fringy player’s perceived value. Why do we all know Anthony Gose’s name? It’s not because he ever looked like a starting quality outfielder. He happened to have a good April once. I saw him traded in industry leagues for non-trivial talent. Remember that year John Buck was a thing?
Players like Gose and Buck are extreme examples. There are plenty of regular major leaguers who will happen to perform well in any given April. Those players will have more trade value for the rest of 2016 (and possibly beyond).
Owners who build an early season lead probably own a few of these players. If they’re savvy, they’ll sell high for a more reliable player. Owners with a lead can better afford to manage downside rather than upside. Maybe Buck keeps bashing like a 40 homer star, but he probably doesn’t. Why not deal Buck and a second piece for steady Yadier Molina (remember, we’re talking about 2013 here)?
This phenomenon happens for a few reasons. While the player is still hot, it’s possible to fall for his “breakout” as real. We’ve all done it. Even once the player regresses to his normal production, it’s tempting to view the fiery period as repeatable. Often, it’s not. Further, trades are at least casually weighed based on performance to date. The fast starters have superior stats. Your brain will intentionally mislead you to expect more of the same numbers over the remainder of the season.
The Solution
I occasionally play a 5v5 team game. It has almost nothing in common with fantasy baseball (although it is an esport). One thing it does have is game balancing. The game favors the underdog using a wide array of methods. My proposal for fantasy baseball game balance isn’t nearly as aggressive.
The answer is weighted monthly scoring. It’s a purely roto-based solution because roto has the biggest issue with early season leverage. As an added bonus, this scoring system should also keep owners attentive through the entire season.
Here’s how it works. Break the season into one month periods such that there are six segments to the season. Calculate monthly scores for each category. It might look something like this:
| Month | Apr | May | June | July | Aug | Sept | Total |
| Runs | (1)*3 | (1.25)*1 | (1.5)*6 | (1.75)*8 | (2)*11 | (2.25)*12 | |
| Score | 3 | 1.25 | 9 | 14 | 22 | 27 | 76.25 |
In this example, we have an owner who finished the season strong in a 12 team league. The number in parenthesis is the coefficient (weight) that I used. In this case, I increased the weight by 0.25 every month. The number after the parenthesis is the month rank as a score. In September, this owner finished with the highest total. They bombed out in May. Our example owner scored 76.25 points in runs out of 117 possible over the course of the season.
Feel free to use different coefficients. You can also divide the Total amount by some number. In this example, dividing by six (months) or 9.75 (sum of the coefficients) both make sense. Personally, I prefer the raw total.
None of the sites I use directly support this scoring system. Your commissioner will need to manually calculate each monthly scoreboard. It’s not hard, it just requires careful attention. Export the league raw stats page on the morning of May 1. That’s your April export. On June 1, rinse and repeat. Then subtract the April export from the June 1 stats. Now you’ve created the raw output for only May. Score it normally and you have the finished leaderboard.
Rate stats can be a challenge. For example, Yahoo does not reveal the plate appearances that go into average, OBP, or OPS. Instead, they uselessly show games played. So whichever stat you use, you’ll have to settle on a work around for calculating it. Using hits or total bases is the easiest alternative if you don’t mind reward volume over quality.
On the pitching side of the ledger, most fantasy platforms include an innings pitched column on the scoreboard. You can use innings to derive the correct monthly ERA and WHIP. Just remember that 10.1 innings really means 10.3333.
If you have an excel savvy owner in your league, he or she should be able to make a useful spreadsheet that auto-calculates everything once you copy-paste the stat export. If I happen to play a league like this in 2016, I’ll re-post this article with our spreadsheet attached.
You can follow me on twitter @BaseballATeam
ummm…why?
I dont play with anyone who buys on players like Buck or Gose. Also April games are just as important/valid as September….
this seems silly
I think there is a version of fantasy BB where the standings are determined by season totals. So, not H2H, but if I acquire 800 runs over the course of the season, it doesn’t matter how many I accrued in a certain week. So the April start allows you to see which categories you can sell. After three weeks if I am leading everyone by 15 HRs, but I’m behind by 5 SBs, I know I can swap out a power hitter for a base stealer.
The advantage is you get to make decisions based on events that you know for sure, your home run lead and your SB total.
Other teams that are behind don’t know how their players will perform in future weeks.
How do you know, for sure, that you can swap out power for a base stealer? And why would someone else give you a base stealer at an unfair price/discount?
The objections raised by BlueFountain (and others) are so freaking weak. Of course you don’t know, for sure, how your players will perform going forward. But you do know with certainty that you have an early lead. Those stats empower you to make informed decisions about where you need the most help. Teams that don’t have a lead don’t have the luxury of that knowledge. This isn’t rocket science.
Why should teams that have a lead be penalized then? I guess that’s my question…
Of course they do. They know what categories they’re weak in and can choose to address that.
But, any team that has a decent projection system should have a grasp on what will happen in the future. A lead in April doesn’t necessarily equate with a lead in September.
If i’m winning HRs in April because Chris Shelton hit 10 HRs, i’m not going to go trade Prince Fielder thinking I’m set in power b/c someone came out of nowhere over 30 days….
“I don’t play with anyone who buys on players like Buck or Gose”
You administer some sort of pre-admission test to your potential owners as to how much they buy into hot starts? Neat.
[And also, if one never buys into hot starts they of course miss legitimate breakouts, so there’s that…]
This “problem” — having games in April worth the same amount as games in September — is only a problem if you play with a bunch of people who give up their team if they are in last place in the first month. I don’t see this dynamic in my leagues, so I don’t see the need to juice up the latter months. To me, this seems like a solution to a problem that doesn’t exist.
I also don’t find it hard to catch up when my team starts out poorly, provided my team is actually good. When my team is drastically underperforming, but they have good rest-of-season projections, it’s usually obvious. This happened in one of my keeper leagues last year, and I regressed back to where I should have been. On the other hand, when my upside players and lottery tickets all are not working out, then it’s not a problem of April vs. September, but of the players I hitched my horse to. Many last place teams have both problems (underperforming stars, and poorly chosen lottery picks).
Agreed on all counts. I haven’t found this column’s premise to be at all true in any league I’ve ever played in. If anything, it’s extremely rare to run wire to wire.
I’ve consistently started in the bottom 3 of my 12 team NL only and risen to somewhere in the top 5 pretty much every year. I sit there in April and say “good god, all my players are underperforming, but I know they’re good players, so I’m not going to panic.” And they usually revert to their old self.
Interesting idea. I’m already using excel to manually adjust standings for one league – we play 5×5 with a weighted Saves & Holds category (saves count double). Definitely something to think about.
Why? what will this enhance?
First, I like Excel. Second, I play in 10+ leagues each year and prefer having some variety. Third, looking for ways to incentivize activity throughout the entire season is a recurring theme in fantasy baseball and this format might do just that.
Interesting approach, but why not take it to its logical conclusion and redraft monthly? If the desire is for balance, actually wipe the slate clean each month. The player with the highest average standing over the six months wins.
Otherwise, it strikes me that the underlying solution is to switch from redraft to dynasty and roto to H2H with a seeded playoff at the end of each season. That of course has other tradeoffs, but it is remarkably balanced in terms of the various strategies and tactics available to specific owners in specific circumstances.
THIS. I play in a 16-team mixed keep 12 H2H league (6×5 categories w/ OBP extra) and it is the most active league for trades that I play in and much like real MLB half the league is in it until the end.
Key things to do to make this work:
-2 week long H2H playoff periods
-Start the playoffs in mid-august to avoid the last week of the H2H championship happening in the bizarro last week of the MLB regular season.
H2H gets a bad rap but if you have seeded playoffs and 2-week long playoff matchups the “best” rosters still usually win. There is also wonderful strategy to be had not only in roster assembly but also in how you line up your pitching/batting to counteract your opponent’s strengths/weaknesses.
Thanks. That sounds like a cool setup. That’s what is fun about these kinds of posts, hearing people talk about different approaches they take to various leagues and issues.
“This isn’t a problem in my league” is the Rotographs equivalent of “I don’t know ANYONE who voted for Nixon!”
If you’re such a fantasy vet, you know darn well how much frustration early quitters cause in the game at large. Especially to other devotees such as yourself.
Interestingly, this is not the reason that Brad gave for this format. It’s just the reason that I think actually has merit. It’s not clear if Brad would argue this is one reason why he likes this format, but I am guessing he would; that seems to be the subtext behind “it’s hard to come back from an April deficit.”
His actual arguments for why this is a nice format were not very compelling to me, which is why I offered the alternative argument in support. His arguments seemed to support the idea that an un-adjusted format would yield more trade opportunities, which I would have viewed as a positive!
Agreed on the latter point. I guess I took it the wrong way when you said, “[this] is only a problem if you play with a bunch of people who give up their team if they are in last place in the first month. I don’t see this dynamic in my leagues…to me, this seems like a solution to a problem that doesn’t exist.”
Brad did point out that “this scoring system should also keep owners attentive through the entire season.” So let’s just raise a drink to more trade opportunities!
I don’t buy the assumption upon which this piece rests. April does not count more than other months, and the idea that big category leads or even standings leads early in the season are very hard to overcome is a fallacy employed by bad fantasy players. Good fantasy players take advantage of that fallacy.
In my long term roto league, we call the April lead the kiss of death. No one single april leader in 13 years has ultimately won the championship. Big early season leads seems to freeze those owners a bit causing the “not broke don’t fix” recency bias.
My league is similar, although I think we’ve had a 1 or 2 April champs win the whole thing over the past dozen years.
I would imagine the odds of the April leader winning are slightly higher than 1/N
Part of the problem of this is the same reason Head to head is a dumb format for baseball: a month is too small of a sample size to determine actual team ability. So my team could have a .350 BABIP in September and that is worth double than my crappy month in April? I agree that it maintains interest for teams longer but it does a much worse Jon or emulating a real life league as a result. This is a non starter for a keeper league. Interesting thought tho
The attitude in the previous thread was that fantasy leagues don’t have to (and indeed shouldn’t) emulate real baseball.
I disagree with those people, but don’t make the same mistake I did and assume fantasy baseball players prefer their leagues to reflect reality.
Why does a head to head league depend upon one month? Don’t most leagues either use the record over the whole season to crown the champion or use the record over most of the season to seed the top teams for a playoff that crowns the champion?
I prefer H2H points leagues. I know it’s satanic to all the traditionalists, which i’m fine with. It works great for my league and it’s members. No league setup anywhere will completely thwart early quitters, which ruin the league for everyone, but the points format in my experience tends to extend the life of the league much more than a roto format.
Points leagues also give you the flexibility of assigning higher weights to the categories that you prefer. Really helps alot with relievers. 3 points for holds and 5 points for saves. So an elite 8th inning setup man in alot of cases is more valuable than a garbage closer. So those guys are very relevant in a points setup, whereas in a traditional 5×5 roto league, they’re not even considerable for being drafted.
I disagree with part of this article’s thesis (namely that H2H necessarily suffers from April surgers just like roto since most H2H leagues that I have played in use a seeded playoffs at the end of the season).
However I have to express how much I appreciate content of this sort. Meta thoughts about fantasy baseball, league structures, strategies, and what’s “optimal” spark really good discussions in the comments and make me consider the games that I play from a different angle.
i’ve never been in a league where anyone was actually able to “sell high” on a marginal talent. i think “sell high” is just something fantasy analysts talk about. i don’t think it is a real-life phenomenon.
There was a market for Hanley Ramirez early last year! He was SS eligible, with 10 HRs!
Selling high on John Buck doesn’t mean landing Clayton Kershaw. But if you traded Buck for any kind of core player, you made out like a bandit.
Take Jose Iglesias last season. He ranked 16th by April WAR. If you had offered him for say Brandon Phillips, you probably would have had a deal. And made out like a bandit in the process. Phillips seriously outperformed Iglesias in four category and matched his batting average.
While your statement is true, I don’t see how this format would prevent Iglesias’s owner from hyping him up based on early season stats to sell at a premium. Just because the scoring is different, he’ll still be touted as the player with the 16th highest WAR at the time he’s traded, and his owner will still make out like a bandit. It just means that they guy who trades for him will be doubly punished, because he’ll get all the regression at weighted higher than April stats.
RC, that sounds like a very good point. I’m curious what Brad has to say on that front.
I see that by focusing on side benefits, I distracted from my point. The April stuff and trading is of secondary and tertiary concern respectively. The main point of game balance is to make comebacks easier. That’s what this format is, it’s a comeback enabled roto league.
You can’t stop people from making bad trades or incorrect guesses. This format is not intended to address that issue. I brought it up to point out an inherent advantage enjoyed by April leaders.
Agree with Brad. Selling high means selling based on a raised expectation of future performance. It doesnt mean trading a nobody with a good week for an all-star.
Sometimes you think you’re selling high and the guy just has a breakout season. Blackmon, Arrieta, Matt Carpenter, Brian Dozier, Anthony Rizzo, Wade Davis, come to mind. They made a leap in performance and were able to sustain it.
A small percentage of owners do fall for it, but the vast majority do not. In fact, many shy away from hot streak players entirely out of the gambler’s fallacy that those players will not only regress but somehow be even worse than expected to make up for their hot start.
For me, in a points league where this “advantage” does not play so much it seems to be over-reacting if a manager trades for that fast starter… and I’ll exploit. I’ve seen the exact strategy move in roto leagues he describes, and I’ve always thought it cheapens the final results. Al that said… one should know this in building a roster and measuring your competition.
Nice try, Mr. Johnson, but breaking the baseball season into six separate scoring segments still won’t make the “daily fantasy” gambling sites any less illegal.
Whilst there is likely to be an element of truth when it comes to having a hot start with batting stats, I feel that the opposite is true with pitching. Most likely, they have accrued wins and K’s by having too many starting pitchers on their roster, and have burned through their innings count by August or early September. They then tumble down the table as a result.
The winner of the pitching categories is the one who generally has the highest average of all categories, including wins and K’s, regardless of where they are in April. Quite often I have been last or near to last in wins and K’s, sometimes until June or July. I then start to creep up the standings in the last three months, often finishing first or in the top two in K’s and wins (I spot start most pitchers). People who have the hot start tend to be quite inefficient in their average of the counting stats, particularly K’s.
By having only two or three starters on my roster, with about 5 or six RP’s, I have plenty of roster spots for hitting (I platoon like mad), so I tend to use most of the batters’ 162 game limit. It’s a combination of tactics that have worked really well for me over the years.
A) This idea is needlessly complex.
B) It doesn’t solve the problems it sets out to.
C) You don’t make a compelling case that these things are significant problems.
D) It ignores other problems that the idea would create.
If you REALLY want a solution to stop crappy owners from jumping on hot starts, why not institute a trading start date of June 15th or so.
A – it’s not really complex. You just have to remember to do an export once a month and 5 minutes of excel work.
B – The problem is “game balance.” Perhaps stating that with a title wasn’t clear enough. This scoring system does solve that problem by making comebacks easier.
C – I don’t see a reason why you need to be compelled. Either the alternative format speaks to you or it doesn’t.
D – The main problem is wonky September stats which you could address via the coefficient (if you believed it to be a problem). The other problem is using a volume stat for hits or TB rather than a rate stat.
Well, it’s not really once a month, I assume one would want to do this daily for the last few weeks of the season.
Also, no, titling it “game balance” does not make the problem clear – especially as this whole idea is to make the game not balanced.
Wouldn’t the trading start date be superior in pretty much every way?
Sounds like the author needs to call the Waaaaaambulance
This is a solution looking for a problem.