RotoGuru Menu
Daily blurbs
Message Forum
Forum Keepers


Football Pickoff

Stat Summarizer
Player Lookup

NBA Schedule
Full Season

Next 4 weeks

FanDuel Hoops
Sortable Stats
Daily FDP

DraftDay Hoops
Sortable Stats
Daily DDP

DraftKings Hoops
Sortable Stats
Daily DKP

Star Fantasy Hoops
Sortable Stats
Daily SFP

Market Madness
Contest Site

Historical Stats
Recap of Entries
Unit Returns

MLB Schedule
Base Advances
Player Lookup

Pitcher vs. Batter
Lineup Dashboard
Daily Live Stats

FanDuel Baseball
Sortable Stats
Daily FDP

DraftDay Baseball
Sortable Stats
Daily DDP

DraftKings Baseball
Sortable Stats
Daily DKP

Star Fantasy Baseball
Sortable Stats
Daily SFP

MFL Site

NFL Schedule
Field Goals
Player Lookup

FireLeague SalCap
Sortable Stats
Weekly pts-Elite
Weekly pts-Std

FanDuel Football
Sortable Stats
Weekly FDP

DraftKings Football
Sortable Stats
Weekly DKP

DraftDay Football
Sortable Stats
Weekly DDP

Home page
Hall of Fame

[RotoGuru logo]
[RotoGuru subtitle]

Introducing the 19th Annual ---
Guru's Market Madness Contest
"The Long and the Short of it"

March 2016

The Basics

The general framework of the contest requires you to buy and sell units. Each unit reflects the tournament performance of one or more teams.

Each of the 64 bracket slots in the NCAA men's basketball tournament will be represented by a team unit. (Technically, these "team units" are fantasy sports derivatives contracts, denominated in GuruDollars, or G$ for short. GuruDollars are the only currency that will matter in this contest. Unfortunately, GuruDollars are absolutely worthless elsewhere. But I digress...)

In addition, there will be a number of "basket units", which provide the performance of a group of teams. There will be 16 seed baskets (each representing all four teams at each seed value), and there will also be a conference basket for each conference that is represented by at least three teams. Each unit will have a stated price - in G$. Entrants must sell short 10 different units. (Selling a unit short is like betting against the underlying team(s).) When you short sell a unit, you collect its price in GuruDollars. These GuruDollars can then be spent to buy as many as 10 units. (When you buy a unit, you're essentially betting on it, rather than against it.) Any GuruDollars that you don't spend you get to keep. Your units plus any residual G$ are your contest portfolio.

Each team will generate winnings or incur charges as the tournament unfolds. If you own a team unit (in the financial world, this is referred to as being "long" the team), you collect all winnings, but pay all charges. But if you have shorted the team, then you must pay the cost of any winnings, although you do get to collect any charges. Essentially, the cash flows for a short position are equal in size, but opposite in direction to those for a long position. Basket units provide the sum of the cash flows for the represented teams. The contest winner will be the entrant with the most GuruDollars at the end of the tournament. (The units themselves have no residual value once the tournament is over, so the only thing of lasting value will be your stash of G$.)

Simple, eh?

Winnings and Charges and Bonuses... Oh My!

Each game in the tournament will result in a winning payment for (you guessed it) the winning team. The amount of the winning payment will depend on the seeding of the two competing teams. The formula for the winning team's payment will be the greater of the following two values:

  • the value of its own seed
  • 17 minus the value of the losing team's seed.

An example should clarify this. Suppose, in the second round of the tournament, a team seeded #1 beats a team seeded #8. The winning team will collect G$9, calculated as the greater of its own seed (1) or 17 minus the loser's seed (17-8). However, if the #8 seed beats the #1 seed, then the #8 seeded team will win G$16, which is the greater of its seed (8) or 17 minus the loser's seed (17-1). Piece of cake!

How about charges? If the loser is beaten by a team with an equal or better seeding, then there will be no charge to the losing team. But if the loser is beaten by an underdog, then the losing team will be charged the same amount that the winning team wins. Using the same examples as before, if #1 beats #8, there is no charge to #8, since that team was expected to lose. But if #8 beats #1, then the #1 team is charged G$16. Anyone who is long that losing #1 team will be charged G$16, but anyone who is short that team will collect G$16. Still with me, or is your brain starting to cramp?

On to bonuses. If the tournament selection committee gets the seeding right, all four #1 seeds should get to the final four. (However, that's happened only once in the tournament's history.) In this contest, if a team seeded other than #1 gets to the final four, that team receives a bonus payment, equal to 10 times its seed value. So, a #2 seed would receive a G$20 bonus, a #9 seed would receive a G$90 bonus, yada-yada-yada. A #1 seed receives no bonus for making it to the final four, though, since #1 seeds are supposed to get there. However, there is still one more bonus payment left, and all teams are eligible. The tournament winner receives a bonus payment (a second bonus, if it's not a #1 seed). The basic formula is the same: 10 times the winner's seed. So, a #1 seed will receive a G$10 bonus for winning it all. But a #2 seed would get two bonuses - one G$20 for making it to the final four, and a second bonus of G$20 for winning it all - for total bonus payments of G$40. If a team seeded #16 wins it all, it would receive G$320 (2 bonuses of G$160 each). But I doubt that'll happen.

That's all there is to it. As each game is completed, your portfolio value will go up or down depending on the game's result and its related winnings, charges, and/or bonuses. If you do really crappy (or is it crappily?), your portfolio value could even go negative. That's not a problem. (You won't win the grand prize, but your G$ credit is good with me. Why not? I can just create more!)


The object is to buy the highest-returning winners, and sell short the worst-returning losers. If you can pick an underdog that wins a few games, or even a Cinderella team that makes it into the final four, that could provide a bigger bonanza than correctly picking a #1 seed to win it all.

Each team will be priced approximately equal to its expected value, based upon historical results for each seed and a double-top-secret proprietary pricing model which I have used to price similar derivatives markets in years past. Former business associates of the Guru will know what I'm talking about. (March was never one of my more productive months from my former employer's perspective.)

Since pricing will reflect expected returns, the best teams will generally cost more than the worst teams (although the seed-based formulas do create some idiosyncrasies). If you want to be able to afford to buy a #1 seed, you'll probably have to sell short some pretty decent teams - or you could also short some higher priced basket units. (--No guts, no glory!--) If you wimp out on your short sales, you won't be able to buy much. Bear in mind, though, that although you do have to sell exactly 10 units, you don't need to buy 10. You can't buy more than 10, but you could take all the proceeds from your short sales and use it to buy one or two units (or none), if that's your preference.

Essentially, the contest rewards those who can select which teams will do the best and worst vs. their seed-based expectations. Most contests just ask you to pick winners. This contest asks you to pick the biggest relative over- and under-performers. If you short sell a #2 seed that loses in the first round to a #15, that could be more profitable than buying a #2 seed that almost gets to the Final 4!

As an example of the reward for performance vs. seed-based expectations, the winning team in the 2015 tournament was Duke. The Duke unit was priced at G$20, and then generated winnings of G$73, for a net return of G$53. That was pretty good. But Michigan State, a #7 seed, was also in the Final Four. As a #7 seed, it cost only $G3 and paid out $G119, for a net return of $G116. A #7 seed making it to the Final Four represented a greater overachievement than a winning top seed.


It wouldn't be a typical Guru contest if I didn't arm you to the hilt with statistical analysis. So I have developed several aids to help you think through your strategy:

  • A table showing the range of historical results that would have been experienced by seed over the last 31 years (since the field was expanded to 64 teams).
  • Some illustrative examples which show the maximum potential values for each seed, depending on the round of exit.
  • A tournament scoring simulator, which will produce the hypothetical scores for any bracket scenario that you input.
  • And don't forget to visit the Market Madness message forum, where you can raise questions and talk through potential strategies with other Gurupies.


As stated earlier, your contest score is your portfolio total of GuruDollars at the end of the tournament. The entrant with the most G$ wins. In the event of a tie, then a series of tiebreakers will be successively applied until only one entry survives:

  1. The entry with a long position in the winning team
  2. The entry with the best net gain on its short positions
  3. The entry with the most G$ unspent at the beginning of the contest
  4. The entry which was the earliest one registered.
  5. The Guru will be the sole judge of the application of tiebreakers. To the extent that the scoring rules are unclear, I will apply them the way I meant them to be applied. Any whining by contestants will be promptly and publicly ridiculed.


I will offer a Contingent Grand Prize of $5,000. If the contest winner has a score within 1% of the best possible score (including basket units), that winner will receive a cash award of $5,000. Obviously, the odds of this happening are very remote. But you never know...

If the Grand Prize winner scores within 10% of the best possible score (but not within 1%), that winner will receive a $500 gift certificate to

If the winner is not within 10% of the best possible score, the grand prize will be a $50 Amazon gift certificate.

A $50 Amazon Gift card will be also awarded to the contestant with the best result excluding basket units. While many entrants like the basket unit concept, some feel that it adds an undesirable degree of chance (i.e., dumb luck) to the contest. So, for those purists out there, I will award a $50 Amazon certificate for the best score for team units only. The net return for all basket units will simply be set to zero for the purposes of this prize award. If you don't include any basket units in your entry, then this will be the same as your regular score. But all entries will be eligible for this prize, as long as the short sales for individual team units are sufficient to fund the corresponding longs of individual teams units. If your team longs cost more than you raised with team shorts, then your entry will not be eligible for this prize. If you submit multiple entries, you might want to try one with team units only, just to improve your chances for that prize.

Each of the above prize winners will also earn a highly coveted berth in the RotoGuru Hall of Fame.

In past years, I have also awarded prizes for several additional GuruPatrons. This year, I am not going to do that anymore. This contest primarily for fun, but has generally been a financial loser for RotoGuru. Itís clear that the availability of GuruPatron prizes does not attract additional entries. So while Iíll continue to offer the two championship prizes listed above, thatís it for this year. Although members of the Guru's immediate family are ineligible for prizes, they can (and will) still compete. And of course, the Guru himself will be competing. But since he is a member of his own immediate family, he can't win prize money either. Shucks!

Entry Guidelines

Entries will be accepted once the official unit prices are posted and the entry form is released, probably not until late Sunday evening, March 15. The entry process is a fully automated, online process. All entries must be recorded prior to the start of the first game on the tournament on Thursday, March 19. You can register anytime, however. Click here to register.

Each person may enter up to three separate sets of picks. (To do this, just register three times.) Entries must include the name of the entrant, and prize money will only be paid by a check made out to that name. So, if you decide to do an "end run" and enter additional times using your kids' names too, you'll have to negotiate with your kid if that entry wins. (The same caveat applies for kids who register under their parents' names.) Suspected abuses of this rule will be investigated, and entrants who are found to have more than three entries will be disqualified from prize eligibility. Screening for excessive entries will not occur until after the tournament ends, and award winners will not be final until announced as such.

If you wish, you may also join a division, which will allow you to track your score against a smaller universe of competitors. Divisions are user created, and may be password protected, or open to all. You may want to set up a division for your family, friends, or co-workers. Joining a division is optional, and has no bearing on prizes. Up to you.

I love this game!

This contest is not affiliated with or endorsed in any way by the NCAA.

RotoGuru is produced by Dave Hall (a.k.a. the Guru), an avid fantasy sports player. He is not employed by any of the fantasy sports games discussed within this site, and all opinions expressed are solely his own. Questions or comments are welcome, and should be emailed to Guru<>.

© Copyright 2000-2016 by Uncommon Cents, LLC. All rights reserved.