What is positional scarcity?
Positional scarcity refers to the different numbers of players available at different positions, so in Draft FPL we’re talking about GKs, DEFs, MIDs and FWDs as the positions we have to navigate. You won’t have had to play FPL for very long to know that different positions have fewer elite players to select, with the striker-pool the most obvious position to point to. What this means is, particularly on draft day, we have to make sure we are valuing those elite strikers properly.
A simple way to take account of this idea is to prepare your draft rankings in tier lists – a popular setup in the draft community. Now this isn’t a style I personally prefer, but the advantage is you can quickly scan and take an available player who occupies the highest tier on your board (some would call this value-based drafting). For example your tier 1 group could literally comprise half a dozen first-rounders. Your tier 2 may be a wider group comprising mostly forwards, midfielders and a selection of elite defenders. It may be that goalkeepers don’t enter your board until tiers 5 or 6 depending on how you set up your tiers.
Value-Based Drafting
Putting this into practice is essentially what value-based drafting is. The concept comes from the US where draft format fantasy football for the NFL is long-established and a national past-time. The theory is particularly useful in NFL as there are a number of positions to fill and choose between on draft day.
Essentially the problem this looks to solve becomes evident after round 1. Round 1 tends to go in a fairly straightforward manner but after this it can be hard to know which position to select at each pick, should I be filling up my strikers, is it too soon to pick up a defender, 3 keepers have just gone off the board do I need to grab one before it’s too late? These scenarios are all covered by the concept of value-based drafting.
The footballguys – supposedly the godfathers of this idea – sum it up nicely: the value of a player is determined not by the number of points he scores, but instead by how much he outscores his rivals at the same position.
The simplest way to start doing this for FPL is to work out what the baseline output is at each position (GKs-FWDs). Now there is some subjectivity and variability in the approaches you can take:
- Value over replacement – how much better is Solanke over the next best striker that would be available in the waivers
- Value over last starter – how much better is Solanke over the worst starting striker in the league (so this takes the baseline up a notch)
- Value over next available – a more dynamic value for drafting purposes, essentially if I take Solanke now, how much better is he than the next striker I could draft in the following round – this can be useful in showing where the drop-offs are and help to create your tiered-rankings
One thing to note here is that this will vary by league size. To take 2 extremes: if you’re in a 6-team league, the calibre of player you could pull out of the waiver pile each week is fairly high, whereas in a 16-team league, you’re rarely able to pick out decent options and so players with comparably lower projections will still hold value.
So for FPL purposes you’ll now have 4 numbers, 1 for each position and you can start to work out each individual player’s value above that baseline number. For example in a 10-team league, I might estimate the 10th best keeper (eg the lowest ranked starter) to reach a season tally of around 115pts. My top rated keeper could be Allison who I project will reach 160pts next season, this means Allison’s value over last starter is 45pts (160-115=145).
For forwards, a 10-team league could in theory start 30 each week, I may have my 30th ranked striker projected to score 65pts for the season. So if I have Solanke projected at 115pts, his value over last starter is 50pts (115-65=50).
So even though Allison should put up a higer tallie than Solanke overall, Solanke’s relative value on draft day is superior and therefore you should be drafting him higher (50 vs 45pts). This can be a tricky concept to get your head around at first, and creating close examples like the one I just did can illustrate it quite well. The reason I’d pass over Allison is because the maths tell me I could wait until the next round and still draft a keeper that’s going to score well for the season. Whereas if I pass up on Solanke, the bottom tier forward options (as we all know) are a BIG drop-off.
Larger Worked Example
Let’s imagine a draft FPL team with just one spot at each position – 1 GK, 1 DEF, 1 MID and 1 FWD. And for simplicity, let’s assume it’s just you and a friend drafting in a 2-person league. You both face off in GW1 and you win the round by 12pts – the player scores are shown in the table below. Let’s also assume in this example, that the way the players performed in GW1 is reflective of how they’ll do next week.
| Player A | Player B | Difference |
|---|---|---|
| GK – 8pts | GK – 7pts | +1 |
| DEF – 9pts | DEF – 5pts | +4 |
| MID – 12pts | MID – 3pts | +9 |
| FWD – 10pts | FWD – 13pts | -3 |
| Player A Wins the H2H +12 |
Now just to illustrate the point, let’s imagine you were going to throw those 8 players back into a pile and re-draft for GW2 (so you pretty much know how well each player will do next week in this made up world I’ve invented). If we were going to select the first overall pick, at first glance you might opt for your opponent’s striker, after all they scored 13pts which was the highest out of all 8 players last week. But in reality, the real value here is the midfielder – although they scored 1pt less for a total of 12, that gave you a 9pt advantage over the other midfielder in the mix. In other words, the midfielder has a higher value and is therefore more likely to lead to you winning match-ups and ultimately your league as a whole.
This is why the concept that keepers should only be drafted in later rounds holds true – you’re not getting much of an edge by taking the 10th best GK over the top one. The above example I gave had Allison as +45 over the lowest starting keeper, translate that over 38 gameweeks and it’s a little over 1pt/week. Compare that to delaying on your final striker spot and you can see how the points will quickly swing the wrong way over the course of a year.
Putting this into practice
- Project xPts for all the players you expect to go in the draft
- Determine your baseline numbers as described above – as discussed already, a variety of ways to do this, the most important thing is that you keep it consistent so you can accurately compare across positions
- Determine the values of each player by subtracting the baseline score from the individual player’s projected score
- Now sort players by their value score
- Work out the average draft position of each player – this is where Draft FC comes in! If you have a high-value player who is consistently being drafted in round 8, you don’t need to rush to take him in the 4th round – instead take some other valued players and come back for this selection in the 6th or 7th round.
- Be able to adjust value as your draft develops
Issues with VBD
- The merits of this system are built on the accuracy of your projections – if you’re generally out in these estimations, this system will not pay dividends come GW38. There are plenty of freely available models if you’re not up for doing this yourself.
- Overly statistical and doesn’t take into account how balanced your team is or the real life feel of the draft as it plays out – this is why tiered lists are more often used because they’re not overly prescriptive and if you decide 2 LIV players is enough for your team, you can opt to steer clear of any other LIV players that are currently in your higher tiers.
- Won’t take into account fixture planning, particularly when considering who you want in the later rounds to go into your GW1 line-up.
- Won’t suggest value in hand-cuffing players, eg GK + back-up from the same team, which is of course a very legitimate draft day tactic, especially in larger leagues.
Conclusion
Understanding positional scarcity is key to a successful draft day and subsequent season. If you choose to go all in on a value-based draft strategy or at least adopt it in part that’s up to you, but hopefully this will change the way you think about the process.