Fantasy Football Preview: Christmas come early?

Despite all the magical thinking I could muster, it has been a dreadful week for me in the HitRowZ Championship.  I scored 22 points, the only positive, and I am really clutching at straws here, is that it is a pleasingly palindromic score.  To put my score in perspective, it was exactly half of the week’s average score (another palindrome 44) and a staggering 93 points behind the games overall top scorer for the week Carlo Willemse.

Carlo’s cheerfully titled “Mother F***** FC” managed a fantastic 115 points, 71 points higher than the average and placing him 1st out of 2.9m players.  In doing so he used the popular, attack heavy, 3 – 4 – 3 formation and benefitted from a fearsome frontline of Suarez, Sturridge and Lukaku.  It is Carlo’s tactical palindrome which stimulated my key question for this week; what is the best fantasy league formation?

Learning from the best

Carlo could theoretically have been bettered this week by The Dream team, and whilst no-one actually picked the 130 point Dream Team masterpiece, Carlo was pretty close sharing both a frontline and a 3 – 4 -3 formation.  Incredibly the Dream team score is calculated without benefitting from the Captaincy bonus.

If I look slightly closer to home for success in the HitrowZ Championship, the top scorer this week was Ben Harrison, 92 points (more than twice the average score) and a traditional 4 – 4 – 2.  So it would seem that whilst 3 – 4 – 3 looked like the strongest tactic for the week, it was still possible to score very highly whilst swapping a forward for a defender.  But this quick analysis only looks at a single week, and whilst, as everyone knows, that is a long time in football, it is hardly representative.

If we consider the Dream Team for the full 6 gameweeks to date we find that the less obvious 5 – 3 – 2 formation is leading the way, and in terms of real players and not fantastical dreams it is a 4 – 3 – 3 that has helped Mark Burke to the fantasy league summit.

From all the evidence above it seems like success is not limited to one formation, and of course managers are free to chop and change from week to week, but is there any further analysis that can help steer me to a tactical choice that is at least likely to be more successful and help me make up the ground I lost this week?

Checking the stats

By grouping players into the three outfield positions we can consider if there are on average more points to be gained by defenders, midfielders or forwards and also at what price those points come in terms of player investment.

Looking at the total scores of the top 30 players in each of the three outfield positions we can see that despite the obvious focus on expected high-scoring forward players like Suarez and Sturridge, after 6 weeks of gameplay it is the defenders that lead the way averaging 29 points each,  the top 30 midfielders average 25 and the forwards average only 22.  This trend remains true if we use the median instead of the mean (27, 22.5 and 19.5), suggesting we are not seeing the results distorted by a couple of high or low scoring individuals in the groups.  For those statisticians amongst you the defensive group show a markedly lower standard deviation than the other two groups as well, suggesting picking anywhere in the top 30 will garner a good points score.  In comparison, the 30th best striker has scored exactly half as many points as the 30th best defender and therefore your ability to pick players as close to the top of the cohort as possible is much more important with strikers.

Based on this analysis the Christmas tree formation of 5 – 3 – 2 would appear to have some logic behind it; strengthening at the back where average scores are highest does not appear to be the anti-football one might expect.  In fact, not only does this garner more points (if you can pick within the top 30 players of course!) but it offers more value too with an average price of £5.54m for each defensive player compared to £6.65m for midfielders and £7.86m for forwards.  That translates to 5.24 points per £1m for defenders compared to only 2.81 for each £1m spent on forwards in the top 30. This kind of Moneyball analysis makes it seem obvious where my money should be spent.

Key to making the most of this analysis is of course being able to spot who the top 30 players in each category are.  Assessing the popularity of the current top 30s shows that top 30 defenders are chosen by 9.94% of fantasy players on average, for midfielders this rises very slightly to 10.07% and for forwards the lower 8.75%.  What does this mean?  It seems to suggest we are equally good at spotting the good players in defence, attack or midfield.  There is not a wisdom-of-crowds result here showing everyone already knows where the points are to be found.

Interestingly if we exclude the top two from each category the average popularity remains roughly the same for defenders, but falls by  about 2% for midfielders and forwards.  This suggests that at the very top of the attacking game we really do know what we are doing.  Everyone can spot the headline grabbers.  Over a third of players know enough to know Yaya Toure and Aaron Ramsey are likely to score well.  The same can be said for Sturridge and Giroud. However this certainty plateaus pretty quickly; 3rd place Aguero is only trusted by 9.3% of players, 20% less than 2nd place Giroud and only 12% of you spotted Hull Tigers’ Robbie Brady in midfield.

Whilst you congratulate yourself on your Graham Carr like scouting ability, having selected Daniel Sturridge for more than a week, think about this: one in three of your opponents also have him playing for their side, therefore your net gain over the competition is… zero.  At the same time only 2.6% of players had chosen Andreas Weimann (12 points this week and 9th best forward over the season so far), and with Benteke injured, it is the players who noticed Andreas who might be getting a leap on you.

I checked the week 6 points totals as well as the cumulative scores and the results were skewed away from the general trend towards the midfield.  However, forwards remained the least value (measured in points per pound) and had the lowest median score (3 compared to 6 for defence and midfield), their mean average having been distorted somewhat by the top two performers.  However, sceptics amongst you should note that this is a week where only 3 teams from 20 kept a clean-sheet ruining many defenders chances of contributing to the fantasy points tally.  Compare this to 6 teams the week before, 7 the week before that and 9 the week before that and you might consider this week the exception, I expect a regression to the mean over the coming weeks.

Lessons to take forward

What have I learnt?  Well, to my mind value is found at the back.  Investing in five good defenders is likely to garner a significant points base from which to build your success.  The risk of doing so is lower too as if we look at the standard deviation (the closeness of most players in the group to the average score) it is only 6.3 points, whereas for forwards this is a much higher 8.5 points, suggesting that you have more chance of gathering a good points total from your defensive selections and are less likely to suffer from the risk of not managing a pick from the top 10 players.

For forwards, the big points are shared between the few.  For me then it’s simply a question of not falling behind the opposition by ignoring the obvious. If all of my competition has two of the top five players I would be a fool not to follow suit.  Given Giroud’s contribution to date I think it only fair to stick with him and then add either of Suarez or Sturridge as the chance of a proper strike partnership forming looks good; if one scores there is a high probability the other will pick up an assist. Given the limited return in terms of points per pound though, I am content to only spend money on two good forwards.

In the midfield my key strategy is to chop and change.  This week only 0.3% of players selected the second highest scoring midfielder.  Clean-sheets and goals may be predictable, but spotting form and flair in the centre of the park is where the skill comes in, and where the difference can be found over your competition.  Next to no-one has invested in players from Fulham or Stoke (ignoring the ever-popular Shawcross and Begovic) and with the two teams playing each other there must be some midfield gold to mine that others are ignoring.

I’m going to hope my early Christmas tree reaps rewards and eschew the three at the back accepted wisdom that is populating the top of our league, who says you can’t argue with success?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: