Friday, 6 January 2017

Ten Things I Learned In Fantasy Football

This year, for the second year, I joined my workplace Fantasy NFL Football league, even though I'm nowhere near my 'workplace'.  I work from home in the UK, and most of my colleagues are based in Texas, so I don't get much chance to engage with them outside of a work environment - so I seized this opportunity.  Some of my colleagues asked me if I knew that this was American football (some of them with more sarcasm than others), but they were all very welcoming.  And I can assure you that I know enough about football (I'm going to call it football instead of American Football - it's just quicker to type) to understand the rules of the game, the aims of the game and the basic stats (yards, passes, interceptions and so on).  We use the Yahoo fantasy football scoring system (points per 100 yards, typically, with extra points for touchdowns) - which I soon got to grips with (and produced my own Excel spreadsheet to identify the good players, as you do).

Now, although I understand the rules, I had no idea about who the best players were, so I really did start from scratch - reviewing the previous year's data and rankings, understanding how Yahoo scores each player, and so on.  This means I had no preconceptions (also known as 'experience') about the best players or the most successful teams. They are all just names to me.  Le'Veon Bell's arrest for drugs; Cam Newton's Christian faith (and his fashion sense); Derek Carr's philanthropy... I wasn't aware of any of them.

However, here's what I learned:

1. Some Americans are extremely competitive. Not just the actual football teams and players, but my workmates - and some of them take this very, very seriously.  (I have the advantage of having nothing to lose - after all, is an Englishman supposed to know anything about the NFL?  Don't English men just drink tea and play cricket?).  I had heard about trash talk, but now I know what it means - and thankfully last season, most of it was directed between other players.  This season, there was almost none at all.  Perhaps my American colleagues just weren't trying hard enough?

From "If Brits Played American Football" YouTube video.

2.  As they say when advertising risky financial products, previous performance is not really an indicator of future performance.  It's okay to review a previous season, or even a previous game, but it's not going to give you all the answers.  It's good as an indication of a player's abilities and potential performance, but it's not comprehensive or totally reliable.  More detailed information about player form and fitness, and the strength of their opposition is also important. Fitness levels are important -more than just the "Questionable" that Yahoo listings provide:  wider reading is recommended. For example, Derek Carr (QB) scored 47 points one week... and just 7 the next.  I bet you didn't see that coming.


3.  Yahoo's own points projections are unreliable at best.  I suspect they're produced at the start of the season and not adapted or updated based on circumstances or form throughout the season, because there have been times when my players have massively outperformed them (Le'Veon Bell (RB) and Julio Jones (WR) are two examples) and yet they've not seen their projections change for the following week.

4. MVP (Most Valuable Player) can also stand for Most Variable Player.  I had Cam Newton (QB) on my team last year, and again this year.  I also drafted Derek Carr (QB), who has had a season of two halves.  There was even a week where I played Marcus Mariota (QB), (which worked out for me).


Overall, Derek Carr scored 328 points, 14% more than Cam Newton's 287.  However, Derek Carr was less consistent:  his maximum scores were 47 (week 8) and 31 (week 4), and his minimum scores were 7 (week 9) and 4 (week 14).  Yes, just 4 points.  His overall spread of results is 4 - 47, which is 43 points.  For Cam, the maximum scores are 40 and 26, the minimum scores are 12 and 13, and his spread is just 27 (compared to 43 for Derek).

So, who do you pick?  There's considerable variation in both players:  Derek scored 328, Cam scored 287, but if I'd picked the better player each week (retrospectively), their combined score is 418.  This game is not just about drafting good players, it's also about playing the best one on a week-by-week basis.



How are you supposed to forecast the performances in weeks 8 (47 points) and 9 (7 points)?

5.  I have to pick my draft selections in advance, as I'm six hours ahead of my Texan friends and the draft session is too late in the day.  This is not a significant disadvantage (nor am I complaining), but it does mean I have to choose my list all at once, without knowing which of my first picks I drafted successfully.  It's a lot like running an A/B test (and I have treated the whole Fantasy Football thing like a series of A/B tests) - you have to set up your recipe before you start running the test!

I should probably confess that in my first year, I didn't realise in NFL that you can change/transfer your players each week (it's not like soccer, where there are specific transfer windows) and hence I drafted two kickers - a lead kicker and a substitute.  I didn't make that mistake this year.


Yahoo gets all sassy with my team selections in my first season
6.  It's okay to make transfers to change your team - like I said, this is really just iterative testing with more noise than usual.  It's frowned upon (halfway through the first season, I received the "Most OCD Manager" for the most roster changes), but not against the rules.

Partway through this season, I picked up Jay Ajayi (RB) and Marcus Mariota (QB).  Marcus is the new quarterback for Tennessee - this was his second season - but a few weeks into this season, I noticed his performance based on, and drafted him and played him once.  Similarly, Jay Ajayi has really developed this season, and very quickly became my second running back - my first running back slot was taken since I discovered Le'Veon Bell last season ;-)


7.  It's okay to use the wisdom of the crowd.  There are sites which compile player rankings from multiple sites and will enable you to compare one player against another, week by week (taking into account effects like injuries, opposition, and so on).  This is extremely useful if you have two players in mind - either both players in your squad, or one that you own and one that you're considering picking up as a free agent.  My personal favourite is FantasyPros.com.  These compile the rankings from sites like Yahoo, but also take into account expert rankings which are updated and reviewed every week (unlike Yahoo, as I mentioned in paragraph 3).

8.  Le'Veon Bell (RB) is an extremely good player.  He was suspended for the first four weeks of the season (as I discovered after drafting him) but has still been one of the highest scoring running backs this year.  In week 15, he achieved 55 fantasy points, which was just over a third of my team's total for the week, and possibly the best for a RB in the whole season.

9.  Bye weeks: after a few introductory weeks, each team has a week off during the season, so you'll need strength and depth to carry your team when your best player(s) are not playing.  And it helps if you can stagger your team's bye weeks, so that you don't have a large number of players out in the same week - as I discovered last year, and then remembered too late this year.  This year, I didn't pay enough attention, and had a week where two or three of my best players were all out at the same time.  A note that bye weeks are not the same as in the English Premier League, where an International weekend means that nobody plays.

10.  It's not great when you have players in your fantasy team who are playing directly against each other in a given week.  Are both players going to have good weeks, if only one of them can score points when they have possession?  This is also important when you pick your defence - it's really not a good idea to have your quarterback play against your defence - only one of them can do really well.  And if you're spelling defence with a 'c', and stressing the second syllable instead of the first, 

11.  Yes, I'm having 11 lessons, because the article title is as accurate as a Yahoo player projection.  Lesson number 11 is that if you win, you become the 'commissioner' for the next year.  From what I can tell, this is a thankless task, where you set up all the parameters for the season (the points awarded for yards, touchdowns, field goals and so on) and how many teams make it into the playoffs.  Do it well, and nobody notices.  Do it badly (or less well), and everybody complains, especially at the end of the season when everybody claims they've won; that they scored the most points; conceded the fewest; made the most player transfers (I thought this was a bad thing, but apparently not); and won in the playoffs (which everybody, for some strange reason) was entered into.  Our commissioner this year did a great job.  That's all I'm going to say :-)
My results?  I achieved 6-7-0 for this season, making the play-offs by coming third in our league of eight, and then coming third in the play-offs  My aim was to be not-last in our league, and I exceeded my own expectations.  I even made some of my colleagues nervous by winning my first two games, and climbing towards the top spot.  My weekly points average was 128.62, with a high of 192 in week 10 (Le'Veon Bell 33, Stefon Diggs 31, Cam Newton 26, Julio Jones 25) and a dreadful low of 66 in Week 3 (Willie Snead 0, Julio Jones 2).

Next season: I'll read in advance of the start of the season to identify any suspensions or injuries, then review the best players from this year. My spreadsheet is ready!

No comments:

Post a Comment