Learning From Previous Bets

As we close out the first Quarter of the baseball season, I noticed that I hit the magical 5000 frequent Player Points mark, which means I'm halfway to getting my full initial deposit reimbursed.  I've played a lot of games to get there (889 to be exact).  I think it's enough of a sample to look back and try to decipher what seems to be working and what's not.  But first, I'd like to make this statement, which is true, misleading, and makes me seem better than I am at Daily Fantasy.

I deposited $100 at the beginning of the season and I've already won $1217

This makes me think of the ads for Daily Fantasy Sites.  While it is true, it doesn't take into account the slightly over $1000 I've put at risk and lost.  I'm really only up $83, once you deduct my wagers and money earned via matching deposit bonus.  I just wanted to point this out so that you can view those ads with an appropriately suspicious eye.  It's kind of like being a Stock Daytrader.  Done right you can show a 1099B with well over $1,000,000 on it because it only reports sales.  It makes no mention of the well over a million you used to buy the stocks.

Anyway there are several styles of games, and I thought it might be good to look at them a little more closely,  to see what there is to learn.  The numbers will add up to less than 889 because I'm only looking at true tournaments, Head to Heads, Double Up and Triple ups, and not the very small group of strange and often different in-between games.  My overall winning percentage is .509, but here's the results for these types of games:

	Tourna-ments	H2H	Double Up	Triple Up	Totals
Wins	27	281	42	52	402
Losses	63	229	46	79	417
WPCT	.300	.551	.477	.397	.491

The worst percentage is the tournament number at 30%.  My thinking is that one should pretty much stay away from tournaments as much as possible.  The lure of the big bucks is attractive, and probably what got me to enter so many.  Admittedly, over half of these games were Quarter Arcade (47/90), where I was just feeling out how tournaments worked.  It's easy to think from a profitability standpoint that one big hit will make up for the 7 or 8 losses out of 10.  But it doesn't if you look closely; in only 7 out of the 90 tournaments played did I win at a greater clip than three times my investment, with 6 paying 3x and 12 paying 2x (there were a couple that due to ties at the bottom won less that 2x).  Considering that the likely outcomes for tournaments say that one will win double your money or less greater than 80% of the time, one is probably better off using the higher dollar entry fee to compete in a $10 head to head or Double up.  Try not to fall too much for the allure of easy money, my biggest return ever was 6x the entry fee, players have to score in the top fraction of 1%. score really good returns [in the $1,000,000 payoff pitch 80% of people will completely lose their $27 entry fee; 13,3% will double their money, 5.2% will triple; meaning that only the top 10% will score better than other types of contests; also only 2/10 of 1% will win more than $500, the kind of money that makes you consider the game in the first place].  Anyway, the lesson is that tournaments should be entered very sparingly.

The second worst winning percentage is the Triple Up, but because you win at least 2.7x your initial entry (some of the triples are three team leagues), a 40% rate yields a small margin of profit.  I like to think that when you put together the sort of roster, you feel is tournament worthy, it's a better bet to attempt a Triple up or create a three team league.  If you're looking for a bigger payout, this type of game is the only one that can be considered mathematically worthy.

The Head to Head and Double up numbers are the most interesting.  I have an above .500 record in H2H's and a below .500 mark in the Double ups.  Functionally, they are pretty much the same thing: same entry fee, same basic cash out (some of the site run 50/50's payout a true double up, as the take the house share by having a few more teams to cover the vig, meaning that more than half the entrants lose).   Where does this discrepancy come from?  Most Likely, it's because a lot of the really skilled players enter the GPP double ups and 50/50's, and if there's 10 winning places they nab 3 or 4 winning spots, leaving the casual player in a situation where they need to be in the top 6 of a 16 team field, making the match work more like a triple up for less money.  If you're going to play in a 50/50, you're going to be better off playing in a player created, non-guaranteed match over the one's the site creates.  I know I've finished just one or two slots out of the money in a GPP double up an incredibly frustrating amount (22 of 46 losses).

As long as we're talking about the quality of competition, a closer examination of the H2H games reveals something quite interesting.  There are two ways to get into a head to head match:  you can accept an existing challenge, or you can make the challenge yourself (Draft Kings calls it "hosting." If you break down my numbers into these two categories, an noticeable trend emerges:

	Hosting	Accepting
Wins	138	142
Losses	92	137
WPCT	.600	.509

As a Challenger, I win about half the time; but as a host I win nearly 1 more match out of ten.  You can see that the number of wins are nearly the same, but the I've lost 45 more times when I've accepted a challenge.  It's pretty clear, I should host more games, as most of us should.  The lion's share of the hosts of the games are the grizzled veterans of the game; and they are more likely not to make the kind of mistakes that cause losses.  They don't beat themselves: when they get beat, more often than not, they get beat by someone that does well that day, not because they do poorly.  There's a reason why you can see them putting out dozens, if not hundreds of matches out there.  They know the deal.  They want to get at the less experienced player.  Someone new to the site or just testing the waters isn't hosting 10 $5 matches; he or she's scanning the lobby for a game or two to play.  And you're only going to get that game, if you host. It's the bait looking for a fish. Additionally, the truly top tier players, aren't scanning the lobby for a $1 or $2 game to join.  They are putting $1000's at risk daily, the effort to join 100 bucks worth of buck games isn't worth it.  However, they are not opposed to tossing out 25 $2 challenges or 50 $1 ones.  It seems that being the host brings with it a better potentiality for weaker (at least less seasoned) competition.