Average Score

Just thought I would post some interesting numbers. Hypothesize what you want about them. I'll update these every few weeks as right now they are probably skewed by low counts (Each category has at least 4 instances).

Players	Avg. 1st Place Score
2	44
6	55
10	62

Contest Cost	Avg. 1st Place Score
$1.00	48
$4.40	56
$5.50	51
$11.00	47

Contest Prize Pool	Avg. 1st Place Score
$1.80	40
$5.40	50
$9.00	62
$10.00	48
$20.00	50
$24.00	55

Contest Cap	Avg. 1st Place Score
1500	57
1600	54
1700	41
1800	65

A Few Thoughts . . .

1. It's difficult to set a fantasy baseball lineup on Sunday. So many MLB teams use Sundays as "off days" for regular players - especially catchers. If you're going to play in a league on a Sunday, it's really important to check the posted lineups of early games to ensure someone isn't sitting it out. If you can find a Sunday fantasy lineup that doesn't have anyone sitting out, you're already ahead of the curve.

2. The Brewers' Bill Hall is a stud against left-handed hitters. His line last was night was worthy of an all star: 3-for-6 with a home run, a stolen base, four RBI, and a run scored. Most of this productivity came against Houston's left handed-starter, Mike Hampton. Being from Milwaukee, I'm well aware of Bill Hall's prowess against left-handed pitchers - so I had him active in every league I played. At the same time, when the Brewers face a right-handed starter, I don't even glance in Hall's direction when creating my teams.

3. What happened to John Beckett and A.J. Burnett yesterday? In a game projected to a be a pitcher's battle, the stat lines for these two pitchers was U-G-L-Y. Each pitcher only survived five innings and they both gave up 8 earned runs. If nothing else, the ineptidude of each pitcher allowed them both to escape without being credited with a loss. In Draftbug, Beckett earned -4 points. Ouch! Burnett barely did better, "earning" -1 points.

'Table Selection' For Daily Fantasy Baseball

Pro poker players sometimes talk about 'table selection'. Basically, the idea is to avoid playing against players who are better than you, and seek out players who are worse. In most cases, table selection will have more impact on results than any amount of practice. No matter how good you are, there's always someone better. And no matter how bad you are, there's always someone worse.

The idea of table selection can be applied in several ways to Draftbug and other daily fantasy baseball contests.

The first is very similar to how it's applied in poker. Take notes on your opponents, avoid those who are better players than you, and seek out games against those who are worse. While it's unlikely that you'll have enough contests against most opponents to be able to have statistically significant results, some players will be so weak that you can look at their picks and identify obvious bad picks and mistakes. I know that at least one of our group is taking detailed notes, and it's something that anyone truly seeking to become a professional should do.

The second approach to table selection in daily fantasy contests is something that I've been doing. On days when I don't like the lineups I'm able to create, I'm playing much less heavily than on days when I do like the lineups. I'm averaging about $20 - $25 of contests, but on days when my top rated players are all expensive stars, and I expect my lineup to look just like everyone else's, I'm entering much less. On days when I can identify at least 2 or 3 players who I think are great bargains due to their matchups, I'm playing as much as $30 of contests.

Not a Millionaire Quite Yet

I'm not a millionaire quite yet. I have been doing pretty well though. At least until I went 0 for 3 ($27.50 total entry fees) yesterday. Steve's next post is going to include a summary of how all of us have been doing, but my ROI is still very high. One thing I've been doing is playing more heavily on the days where I like the picks available to me. Usually that means one or more teams have extremely favorable matchups for their hitters, allowing me to construct a nice lineup within the salary cap. Other days (like today) I'm playing in fewer contests and cheaper contests. $33 contests were added to Draftbug's contest 'menu' a few days ago, but so far I haven't had the guys (or the bankroll) to play in any. The highest buy-in I'm playing frequently are the three person $16.50 contests. The challenge with those is knowing whether to treat them strategically as if they were a heads up contest, or a larger multiplayer contest.

Four Days In

Four days since joining Draftbug, I can't say that I've found much success so far. In particular, I haven't been able to select an effective starting pitcher to save my life. Even as I type, the pitcher I selected in most of my leagues this evening (Jake Peavy) just gave up a grand slam to Edgar Renteria. Ouch! During my first four days, I don't know if I've selected a pitcher who has earned a victory. That doesn't help!

As Alex pointed out in an earlier post, he made over $50 yesterday - thanks partly to me. I thought I was looking pretty good when I went to bed with Arizona leading Colorado by 4 runs in the 7th inning and Alex's only hope of victory riding on the right arm of Chad Qualls. Unfortunately (for me), Colorado scored a single run in the 8th inning, giving Qualls the opportunity to earn a save for the Diamondbacks. The points Alex earned from Qualls were just enough to squeak past me in that tournament.

Oh, well. I'll get you next time, Alex!

Thru four days, my bankroll has suffered a hit of exactly $40. Surprisingly, I'm not discouraged. Once I start hitting some points from a starting pitcher, I'm confident that I'll start winning consistently.

If nothing else, this website has allowed me to become reaquainted with Microsoft Access, which I hadn't used in awhile. I'm tracking all my tournament results with this software. What is everyone else using?

Predicting Homeruns in April

So at the beginning of the season I entered into 4 April Home Run Derby’s at Draft Bug for a total of $17 + Freeroll and I haven't really looked at them since entering, so I thought now was as good a time as any.

I’m 1st in 2 of them (4 and 6 person), 6th in a 23 person and 14th in the 200 person Freeroll. Pretty decent considering I had no idea who to take, so all I did was guess who would finish with the most HR’s and pick them. The numbers below suggest that was a terrible strategy.



In 2007 of the top 15 in the month of April, 5 were in the top 12 of 2007. In 2008 it was worse with only Quentin being in the top 12 of both April and 2008. How many HR you have in April has very little bering where you place for total HR. So picking your team based on total season HR is a very poor strategy. I would not be surprised if over the last 10 or so days in April if I began to fall and didn't place in any of them.

My First $50+ Winning Day

With only seven games scheduled yesterday, one of those too early to be included in Draftbug contests, and four of the remaining six threatened by rain, there were all sorts of interesting strategic issues to deal with. I ended up playing in much higher value contests than normal, because I saw what I believed to be an unusually profitable opportunity. The same player had entered a $22 heads up contest, an $11 heads up, and a $16.50 three person contest. He's someone who I believe is knowledgeable, but usually enters early and doesn't seem to be logged in that much during the day. My thinking was that if he ignored the weather report I would probably sweep him away in all the contests, while if he did pay attention to it, we'd still on relatively even footing. I selected only hitters in the two games that were sure to be played, but choosing starting pitchers was a little tougher. Lowe was the clear standout, and I knew the Atlanta game might happen. In the end I used Lowe in the three person contest (where I thought a higher score might be needed) and Arroyo (who I knew would get to pitch) in the other two contest. It ended up working out well as all three of us picked Lowe in the multi, while Arroyo outpitched Lowe. Things were complicated a little when Steve became the third player in the $16.50, but everything still went according to plan as I won all three contests. Total entry fees: $49.50. Total prizes: $105. Cumulative entry fees: $174.40. Cumulative prizes: $275.25. ROI: 57.8%.

Quick Results Update

Here's a quick update on results over the weekend. I entered five contests on Saturday and just one on Sunday. I won a four person winner take all ($4.40), tied for first in a 6 person $1 contest, and won yesterday's contest (which was a 6 person $5.50). Entry fees: $28.50. Prizes: $38.20. Cumulative total fees to date: $124.90. Cumulative prizes to date: $170.25. Return On Investment: 26.6%.

Week 2: Bankroll Management

The lesson learned from Week 2 is the importance of Bankroll Management. I have entered 27 2-Player Contest this year. I have scored under 30 points in 5 of those contests. Of course three of those came on the day I decided to enter contests with higher entry fees.

It is for that very reason that you can't put all your eggs in one basket. Luckily, I only had $22 invested on this day of struggling. Even the most accurate of models will struggle from time to time. Although this was a hit to my bankroll, it was not devastating. The terrible day on Saturday set me back, but here are some of my stats through two weeks of the season:

Week 1 Average Team Score: 54.3
Week 2 Average Team Score: 35.9
Overall Average Team Score: 43.4

Overall Opponent's Average Team Score: 37.9

Outscoring my opponent by 5.5 points per contest should yield profitable results in the long run.

Introducing Steve Schroeder

We're happy to introduce a new member of the Draftbug Millionaire team. Steve Schroeder entered his first real money contests at Draftbug yesterday, and proceeded to crush me in a $7.70 contest that we both ended up in. Here's what he had to say about himself:

Hello world!

My name is Steve Schroeder and I’ll be playing as stv1313 on Draftbug. If you see me online, feel free to say ‘hi!’ Born and raised in Milwaukee, I have been actively playing fantasy sports since 1993. In that year, I played my 1st fantasy football league and took Barry Sanders in the 1st round. It was a TD only league, and Mr. Sanders only scored 3 TD’s that season. I’ve never liked Barry since.

Since then, I’ve played almost every fantasy sport imaginable. Even as I type this entry, I’m an active participant in fantasy baseball, golf, and NASCAR leagues. Apparently, my life is just a fantasy! I also enjoy online poker and I’ll make a wager on almost anything imaginable during my day-to-day life.

Approximately 18 months ago, I discovered the exciting world of online fantasy sports. Playing at another online site, I found great success playing football, basketball, and baseball. In 18 months, I accumulated a bankroll of over $3,500, despite the fact that I never deposited a penny at that site. I’ve used those winnings to purchase a freestanding Ms. Pac Man machine for my wife and the remaining loot has been set aside for a trip to Vegas to play in the World Series of Poker in June. It’ll feel good to play on a “freeroll” while in Sin City!

Unfortunately, my previous fantasy site isn’t offering baseball this season, so that’s brought me to Draftbug. I look forward to seeing if my previous success carries over to this new site. Can I turn another nonexistent bankroll into over $3,500 again? Only time will tell!
I’m looking forward to sharing ideas and feedback with y'all. I’ve already read all of the posts on this site and I can’t wait to contribute. Let’s have some fun, win some money, and enjoy the ride!

Week #2: Better than the First

So after an embarrassing week #1, the second week was much more optimistic. As my ROI table shows.

I had a small but positive ROI and helped reduce my total debt. It’ll probably still take me at least another 2 weeks to get back to par, but that’s what I deserve after being over anxious to start the season. My week one average score was 32.2, while my week two average score was 45.7, a significant improvement.

From here on in I’m going to be referring to my mathematical/statistical system of picking players as “Sharon”. My friend coined the name, and it gives a humanistic touch to what would otherwise be just a formula.

I’ve also decided not to calculate and use freeroll percentage as a measurement of success. It's a bad idea as the freeroll has a different points limit and often includes players from earlier games which I do not pick in my late game. My new target is 45 points every night.



Among a number of other things, this week I learned that fantasy baseball comes down to pitchers and for me that means Jonathon Broxton. Broxton is my rock and my redeemer and now I have a huge man crush on him. Below is my story.

Saturday 11th:
Solid day all round. I went with the Padres combo of Peavy and Bell and it paid off big. With a couple of points from McCann, I would have placed in the Freeroll. Great start to the week.

Sunday 12th:
In an attempt to make sure that I do not take any players who are not in the opening lineup, I decided not to play on Monday. All the games were spread throughout the day, so I couldn't check the lineups.

Monday 13th:
I got caught up in the La Russa blender. Again. I really need to learn. I’ve decided the only Cardinal I’m every playing is Pujols, unless I see the lineup card before the game. I won one of my three contests.

Tuesday 14th:
I am such a slow and stubborn person, Fool me once shame on you, fool me twice, shame on me. Matsuzaka will never have a place on my roster again. I will learn. I promise. I had a discussion with Alex about my pick of Matsuzaka and here are the reason why Sharon decided to take him:

1. Oakland is a terrible hitting team, both this year and last. 
2. He's going to get lots of run support. So even if he gives up 4 or 5 runs, its not a huge deal.
3. Only 6 times last year in 29 starts, and 7 times in 2007 (in 32 starts) did he have 3 or less strikeouts. He averages 5.7 SO a game. Oakland is 9th in the league in SO, and was 4th last year. Sharon figured she was going to get at least 6 from him.  
4. He averaged 5.7 Innings per start last year, and 6.3 the year before. Sharon thought 6 innings from him would be fine. 
Boston had a really good chance of winning that game.

Here are the reasons not to take him:

1. He’s terrible.

Wednesday 15th:
Here where my infatuation with Broxton begins. With Lowe only giving me 5 innings and the Braves giving him no run support, I needed big points from Broxton. So he said, “No problem Kyle” and he came through with the win. Oh Yah, Kinsler had a big game to. Kinsler + Broxton = Money in the Bank

Thursday 16th:
Solid day thanks to the Doc but I played some really stiff competition and did’nt place in any of my competitions. I really like picking Guerrero, but unfortunately I missed the fact that he went out prior to the game with an injury. You may notice that Broxton did'nt get me any points, but this is not his fault. What business do the Dodgers have being up by 5 in the 9th and bringing in Ohman?

Friday 17th:
Sharon really liked the Royals matchup against the Texans, especially for the cost of the players. The Royals put up 12 runs and 19 hits. So that was a positive. Oh yah, Broxton, another 15+ point night and the Save. His season stat line? 6.2 IP 1 H 1 ER 0 BB 11 SO 1 W 4 SV, for a total of 67 Fantasy Points.

High Stakes Fantasy Baseball

Ok...not exactly high stakes. Not yet. That's going to take some time, both for us to build up our bankrolls and for there to be enough traffic to support higher stakes contests. I had my best day yet though. I entered 4 contests (total entry fees $16.40). I won three - a $5.50 heads up, a $5.50 4 person winner take all, and a $4.40 6 person contest. I also came in 2nd in a 10 person $1 contest. Total prizes amounted to $48.30. Cumulative entry fees to date: $96.40. Cumulative prizes: $132.05. ROI: 37%. My pitching was bad again, as Shields struggled and my relievers didn't get into the game, but I was saved by a big Tampa Bay comeback that gave Shields the win and a respectable score.

4/14: Results...Back in the Red

I couldn't help myself. I thought I'd start my hiatus from playing yesterday (since I'll be away most of today and tomorrow), but couldn't resist the juicy matchup that the Texas hitters offered. They managed to knock Simon out of the game after an inning and a third, so my opinion was more or less justified. My pitchers were (yet again) less successful. I used Vazquez on every team, and he allowed three runs and lost despite striking out a dozen batters. None of the various closers I used got a save. I'm going to chalk it up to variance (mostly), but I already knew that my calculations for pitchers aren't as sophisticated as what I'm doing for hitters, so I'll probably work on improving what I've got for the pitchers first. I won a pair of heads up $5.50 contests, and was 2nd in a $3.30 6 person contest. I also finished out of the money in another 6 person contest and a 10 person contest. Entry fees: $18.60. Prizes: $26.30. Cumulative entry fees: $80.00. Cumulative prizes: $83.55. Return on Investment: 4.4%.

4/12-4/13: Results. Variance or Lack of Skill?

Over the past two two combined, I've entered 8 contests of various types (2 heads up, 3 six person, 3 ten person) for a total of $25.00 and managed only a tied for second place in one of the ten person contests. Total prizes: $2.25. Total invested to date is $61.40, total prizes to date are $57.25. Return on investment is -6.8%. For the most part, my chances were killed by the poor performances by Tim Lincecum and Kevin Slowey the past two days. I knew that by using the same starting pitcher in most contests I was taking a 'feast or famine' approach, but felt that Lincecum was such a clear standout and Slowey such a good value at the price that it was worth it. My sample size is very small so far. I think this is variance more than a true indication of my ability to win with my current methods. That said, why leave things to chance. I know of lots I can do to improve my projections, and need to put aside some time to begin making changes. I'll be away Wednesday and Thursday, but will try to put some of the changes into place over the weekend.

Week 1 Analysis: Find your Strengths

Without any prior experience in daily fantasy baseball contests, I've decided to take it slow while I get my feet wet and tweak my approach. However, I would consider week one a success. I've entered 13 contests, won 7 of them, lost 4 and placed in the remaining two. Here is a breakdown of the type of contests I've entered and my record for them:

2 Player Salary Cap: 3-3 , -$3.80
2 Player Live Draft: 4-0 , $8.10
10 Player Salary Cap: 0-1-1, $0.70
6 Player Salary Cap: 0-0-1, $5.00

Overall Week 1 Record: 7-4-2 , $10.00


In Week One I wagered an average of just over $5.00 per day, with a total of $35.80 wagered and $10.00 of net profit. Equaling a return on investment of 28%. Although it's a small sample size, I am 4-0 in 2 Player Live Draft contests. I think it's important for anyone who is serious about Draftbug to keep records. This will allow you to find what game formats you excel at, or even find opponents who are above par.

The 1st Five Days = Rough

I’m not sure yet how I want to present these updates, so bare with me over the next few weeks while I figure this whole thing out.

To summarize the 1st five days as nothing but fantastic would be a complete lie. They were rough, and probably to be expected. My system was working wonders in basketball, but thats because I had a whole season behind me to use as a model for future games, as opposed to in baseball where I’m relying on previous years stats, which do not necessarily predict the next season as well.

Below is a summary of my teams over the last 5 days. The second column after the players name is their cost and the third is the points they received that night. The “Freeroll Percentile” is what my percentile finish in the daily freeroll would be with my team. For example if I placed 14th out of 30 players, I’m in the 53 percentile, implying that if played 100 2 player games I would win 53 of them. The reason I’m doing this is to see how I fare against everyone. I could easily win a single game for $50 with a total score of 20, if my opponents score was 19, but that’s just luck, not skill. I’m striving for about 75% on a consistent basis, allowing me to play 4 games a day and projected to win 3. Finally the +/- money line, just shows how much I’ve bet, won/lost, and the difference between them.

Note: I can't pick the same team in every contest, due to cap limit differences. The teams below are my “primary” teams- the team I use for the draft were I wager the most amount of money, which usually has the strictest cap. My other teams only ever differ by one or two players.

Yah, not so amazing.

Monday 6th:
Pretty decent, not much to comment on. Can’t be happen with Morneau at -1 and Lee at 0, but it will happen.

Tuesday 7th:
LaRussa decided to blend his batting order as he’ll do often this season, and Ludwick got left out. Your not going to win any games with five players with 0 points. I was very happy with Bedard’s 14 point performance for a 10 unit cap hit.

Wednesday 8th:
Another bad day. Kazmir had 79% of all my points.

Thursday 9th:
Simply embarrassing. I nailed Scutaro and Beniji Molina, and then proceeded to crap the bed with Polanco, Guerrero, Wells, Crede, Matsuzaka, and Bobby Jenks.

Friday 10th:
The main troubling spot was that in the first four days I picked five people who were not in the opening line up. So on Friday I made a point of not only checking every 5 minutes for lineup updates, but I also refused to take any CIN or PIT players as I was worried about the possibility of a rain out, which did occur.

I got beat in three games including the Freeroll on Friday due to the extended rainout in Atlanta and Lowe not retuning to the game to collect his win where he had 6 SO in 3 Innings and was leading 3-1. Also some of my sneaky competition had selected R.A. Dickey as their RP even though he was starting and was able to pick up real cheap innings and a win.

By the way props to Kaiseroll13 who destroyed me every day this week. It really didn't take much, but good for him for beating me up.

Friday 4/10 Results - Shawn Hill and Rate of Return

After my day off on Thursday, I played in five contests on Friday - 2 10 player $1 contests, 2 heads up $3.30 contests, and a 4 player winner take all $3.30. Rainouts and rain delays played a huge role in all contests, and overall helped me, as I had steered clear of the games I thought most likely to be rained out, and seemed to be one of the people with no Cincinnati players. While I had Lowe on many teams, his getting pulled after a long rain delay in the fourth inning didn't hurt much, because many of my opponents had him as well. I came in 2nd in one of the 10 player contests (thank you Shawn Hill), and won one of the heads up contests and the 4 player contest. Total investment $11.90, amount won $20.70. Cumulative total so far this season is now $36.40 invested, and $55.50 gross winnings ($19.10 net profit), for a return on investment of 52%. That's definitely an unsustainable rate of return, but still a very encouraging start. To things I've noticed are that most people are not watching the weather report closely enough, and most people clearly haven't read or have chosen to ignore the articles I've written on strategy for multi-player contests. That's part of the reason why I'm very comfortable talking about many of my strategies...I know that even those who do read what I have to say will often ignore it, so I won't have much impact on how tough my competition is.

I probably won't get to play in any contests today, and may not on Sunday either. Have I mentioned how incredibly relaxing and fun I'm finding it playing in a game where I can set my daily lineups to exploit favorable matchups, but can just skip days whenever it's convenient for me?

Day Off

That will be today, for me. I'm very unlikely to get a chance to enter any contests today. One of the things I'm really loving about the daily contest format is the ability to just ignore the games on the days when I don't have time.

As for yesterday, it was a success. I entered three contests. The stupid one was a $4.40 heads up live draft for early games only. The stupid part was that I didn't spend much time preparing, and realized once it started that I had no idea who the highly ranked early players were, and no easy way to sort through my spreadsheet to find those players. I lost.

I was also in a $3.30 6 player contest and a $1 10 player contest and won both of those. It's going to take hundreds of contests (especially in the multi-player ones) to have any idea how we really stack up against the competition, but it's always nice to get off to a good start. The key is to use that as motivation to continue improving the techniques I'm using.

Edwin Jackson?

I had a pretty busy day yesterday, so I only got to enter one contest. It was a $5.50 heads up contest against an unknown opponent. I managed to win, despite my Dan Haren getting outscored by his Edwin Jackson. Unless you're going to play in an extremely high number of contests, you're never going to have a large enough sample size of contests against each of your opponents to be able to tell how good they are based on your results against them. So if you're planning to practice 'table selection', you should simply make a note of any clear mistakes they make. One mistake could just be a mistake (like my playing Napoli for the first two days of the season), so you need to really look for a pattern of mistakes to know that a player isn't strong. My opponent's use of Jackson suggests to me that he likely doesn't rely on the same pitching metrics as I do, but it will take more than that for me to know for sure how he stacks up as a player.

Introducing Kyle Huberman

The final (for now) member of our team is Kyle Huberman. I'm really excited to learn more about his approach to fantasy baseball, since he's got more formal and technical training in some of the skills that the rest of us are trying to learn through experience. Here's what he had to say about himself...

Hey Draftees,

Let me begin by introducing myself as Kyle Huberman or IceHube, as you'll find me at DraftBug. I'm 22, and recently graduated from the school of Business at the University of Alberta with a Bachelor of Commerce in Entrepreneurship and small business. Unlike Josh and Tom, I have much less Fantasy Baseball experience, most of my fantasy knowledge comes from the Hockey and Football pools.

The skills I bring to Draft Bug Millionaire are more academic than based in baseball knowledge. I couldn't name you 3 guys off the Royals, Marlins, or a dozen other teams for that matter. But what I do know are the intricacies of different prediction methods and how to use statistical models to help predict future outcomes. My tools are primarily regression analyses, prediction markets, and using Risk Solver Premium in the matlab environment for optimization questions. Because of this you'll find that my teams are often quirky and often lack the traditional larger names associated with fantasy baseball. I have no idea if this process is going to work right of the bat, although I was fairly successful using the same methods over 10 days of basketball.

I want to thank Alex for giving me this opportunity and look forward to playing against all of you over the baseball season. I'm online virtually all day, every day, so follow me on Twitter (@kylehuberman) and ask me any questions you like. In the next few days I hope to finally get my site up where I'll be publishing my daily picks (after the games have started, sorry!), as well as my results with commentary as to what went right, wrong and what to look for in the the future.

First Week of the Season

The first week of the season will be a time for us to adapt our models, including the fixing of silly errors or omissions. For that reason my strategy will be to enter a very small quantity of contests in the first week or so. On opening day I entered only one $1 salary cap contest, and did not finish in the money. I'd like to thank C.C. Sabathia for a lovely opening day start, earning a whopping -9 Draftbug Points. If I had chosen one of my other Starting Pitchers I was considering (Peavy or Johan) then I would have won the contest, oh well.

Hopefully I can avoid another -9 points from my starting pitcher in day two. Pitchers that look good for Tuesday's matchups are Josh Beckett vs. TB and Dan Haren vs. Col. Some sleepers that you can get for a bargain that I think will have solid starts: Kyle Lohse vs. Pitt, Josh Johnson vs. Fla, and David Purcey vs Det(did I really just say that). I think Purcey can outpitch Edwin Jackson, and has a good chance at a Win, and you can draft him for 10 points.

Opening Day

We'll try to figure out an easy to digest format for posting results in the future. For now, we'll each provide them however we see fit. I entered a total of five contests of various types yesterday. A $3.30 heads up salary cap that included all games. A $1 ten person salary cap contests that included all games. A $1 heads up contest that included night games only. And a pair of $2.20 heads up live draft contests that included night games only. My one big mistake was using Mike Napoli in several contests, despite knowing there was a risk that he wouldn't play. I probably should have factored that risk more heavily into my decision making. I won the $3.30 (which was against a solid player I'm familiar with from Rotohog). I tied in one of the $2.20s. I lost the other three contests (4th in the 10 person). Total investment $9.70. Total prizes $8. Return on investment -17.53%. Luckily the baseball season is a long grind, that tends to smooth out the variance and luck along the way. It's time to start working on a few of the many improvements I've been planning to my statistical model.

Introducing Tom Hammer

Tom Hammer is someone I've know through message boards, emails, and fantasy contests for the past two years. He's had a ton of success at all sorts of fantasy games, and I'm glad to have him on the Draftbug Millionaire team. Here's the introduction he wrote:

Hello everyone, my name is Tom Hammer and I’ll be one of the participants on Draftbug Millionaire. I’d like to start by thanking Alex for letting me participate in this exercise. I’m looking forward to hearing about how other "experts" prepare for the season and the strategies they use to win at these games. I always have an open mind and I’m always looking to improve my own skills. I’ll be playing under the userid "The Hammer" on DraftBug.

Now I’ll tell you a little about myself. I’ve been playing these fantasy games for over 20 years now and have always been obsessed with "beating the system". My idea of fun was coming up with systems to beat horse racing, dog racing, jai alai, and football. I spent hours, days, weeks, and years perfecting these systems. None of them ever became the profit machines that I had envisioned, but they did lay down the foundation of something that did – Fantasy Sports. Using the knowledge that I had gained , I started playing Fantasy Sports for money – and I was successful, very successful. Over the past 10 years I’ve won well over six figures in cash (my favorite prize), trips, TVs, all sorts of electronic gadgets, books, and DVDs. My point isn’t to try to impress you, but rather to show you that making money and winning prizes at Fantasy Sports is indeed possible, if you are willing to use some brainpower and are willing to learn from your mistakes. I don’t win all the time, but I do learn from all of my losses. In that way, losing is actually more important that winning. Losing actually provides the blueprint for your success.

In my posts I’ll attempt to show you my thought process when playing a game for the first time. What things I look for, what simple strategies I use, etc. Hopefully you’ll learn something and hopefully I’ll learn something as well because I know I don’t have all the answers.

Multiple Views

One of the interesting things about this project (and any complex project involving more than one person) is that the participants are going to have a variety of viewpoints, with different history, knowledge, biases, and opinions. As we uncover those, we're going to have to figure out which differences are worthy of discussion in an effort to arrive at a common way of thinking, and which can be don't get in the way of our success. I think Josh's previous post was detailed, well thought out, and was the type of statistically based thinking that I like. I disagreed with much of it.

That's not necessarily a bad thing, for either of us. It's going to give us something to talk about, and will certainly provide an opportunity for one or both of us to learn. The discussion will also serve as a basis for improving the statistical model we're using. Without getting into the specific points I disagree with, I think in general Josh may be making the common mistake of overemphasizing analysis of specific information (ie 'how has Wandy Rodriguez done at home"') that may have occurred somewhat randomly, and unemphasizing general rules based on larger sample sizes (ie 'where does home field advantage come from, and who stands to benefit the most?') I think both of us stand to benefit from learning to properly quantify the balance between the two.

There was actually a really good article about one aspect of this at The Hardball Times recently. Here is the link to

Part 1 and Part 2.

As we work through the different views on things, one of the things we'll need to determine is whether we can be effective using the exact same statistical model, or whether we each need to maintain our own to be effective. I firmly believe that unless you're comfortable with the 'system' you're using for any type of speculation or forecasting, you can't use it effectively. One of the other two Draftbug Millionaires (who will be introduced in the next day or two) has already decided that he definitely will be using his own model for that reason. I suspect that we may each ultimately maintain our own models each with its own nuances, even if we end up sharing very similar basic approaches.

Progress Report

As I type this, Opening Pitch is 3 Days, 3 Hours and 20 Minutes away. Now is the time for Alex and myself to buckle down. Currently we have a pretty basic model set up for predicting player performance, but it will have plenty of room for improvement as the season progresses. In order to succeed in Draftbug, it is very important to remember to analyze each situation instead of blindly following a set of rankings. Here are some examples:

1. Home/Away Splits: There are so many variables that can be the cause of this. Whether it's Park Factor, Being Comfortable or, in Ervin Santana's case, not being able to start the game in the dugout.

Example: Wandy Rodriguez is a servicable pitcher on the road in his career, but he's been downright awful at home. It's important to follow the progress of players like Wandy Rodriguez.

2. Difficulty of Competition: Some players get the benefit of playing in a weak division, which lowers their stats for the season, but they often struggle on the few occasions that they have a challenge.

Example: Jake Peavy has a career ERA of 4.50+ vs. only seven teams in the Major Leagues, six of them are in the AL. You can draw the conclusion that Peavy doesn't fare well against teams he doesn't face often, or that the NL is just vastly inferior... It is very important to look at matchups rather than just the skill of the player.

3. Righty/Lefty Splits: You cannot simple go by a standard prediction, because many hitters are much better suited as platoon players. It is important that your model includes platoon stats.

Example:
Jayson Werth Career Triple-Slash #'s vs. LHP: .291/.374/.545
Jayson Werth Career Triple-Slash #'s vs. RHP: .251/.347/.408

4. First-Half/Second-Half Players: Some players notoriously turn it on after the All-Star Break(Johan Santana, Ryan Howard, Mark Teixiera to name a few). However, not many people look at the other end of this. It is a lot harder to find hitters that finish stronger than they close. Grady Sizemore and Lance Berkman are a couple that fit the mold.

The point I'm trying to get across is that blindly following preseason predictions just will not produce accurate results over the course of the season. When Alex and myself are finished brainstorming, we hope that our model will factor in at least these four variables.

Short Update

With baseball season fast approaching, I've been working on getting my statistical model ready for opening day - updating team offensive projections, player projections, and park factors. Once I'm done with that I can actually start working on improving some aspects of the model that need tweaking - things have been so busy that I haven't had a chance to work on it much recently, unlike last offseason. Which reminds me of another characteristic that certainly helps pros - focus! It stands to reason that you're going to have a lot more success at something if its the ONLY thing you're working on, than if you're splitting your time between multiple tasks. Another way of thinking about this is to think about two people working at the same task - one spends 30 hours per week, while the other spends 60 hours per week? All else being equal, how much more valuable is the one who spends more time? I suspect most people will say 'roughly double'. That's wrong. Not only will is he working twice as much, but he's going to be improving twice as fast! So assuming that the task has some complexitiy to it, after a while his double hours may yield triple or quadruple the results!

Josh and I spent some time chatting over the weekend about how to adjust strategy to live drafts, bankroll management, and related topics. Bankroll management and 'bet sizing' are especially tricky for daily fantasy baseball contests, since you need to figure out to what extent different contests on the same day can be treated the same way as contests on different days. There's no simply answer.