Early Season FPL Price Change Trends and Implications for Wildcard Strategy
Posted by FFPundits
1:47pm Aug 2 2016
This article is based on analysis of data on the number of player price rises and falls per Gameweek over the last two seasons.
Unfortunately, I have only partial data from 2014/15 (GWs 2 to 6) but I captured from (fiso.co.uk) all records of price changes from 2015/16.
These records include the time and date of the price change so it is relatively easy to classify them by GW and manipulate the information in Excel. I’m assuming those reading this have a level of knowledge of the FPL price changes system, and am not going to repeat the basics which if needed, can be found elsewhere.
Most who have played FPL for 3 or more years are likely to be at least somewhat aware of the sharp change in the behaviour of player prices after the 2013/14 season. Up ’till then, price rises happened much more rapidly, with double and treble rises in a single GW commonplace, especially early in the season. Thus the logic of the popular strategy to wildcard early, especially around the first international window and potentially gain significant team value.
The plan being to ‘ride the price rises’ in your wildcard week, banking the 0.1 when a player rose twice and them moving him on before picking your actual team at the end of the week. I personally always felt that the first international window was somewhat overrated, with the same levels of transfers over a longer period not necessarily improving the “profitability” and tended to go the week after where I could last that long. An intention to explode the international break myth was the genesis of my collecting price rise data in the first place.
But it became apparent quite quickly in 2014/15 that something was different, with the popular belief (among people who know about these things) being that wildcard transfers no longer counted towards price changes, as well as possible other tweaks to the formulae by FPL Towers.
As a result of the much slower rate of price rises, many people therefore began to place less value on the “early wildcard for team value” strategy, indeed some suggested that paying attention to team value was now a waste of time with such apparently little advantage to be gained. However, it was also noticeable that the rate of price drops had not slowed, and had possibly increased, and the relevance of this has been rising in many managers consciousness.
I have felt over the last two seasons that (for the serious FPL player) ignoring price changes in the first half of the season was still a mistake, despite the new landscape. I therefore did some simple analysis of the rates of rises vs drops at the start of the 2014/15, hence the first 6 GW data, before deciding to capture the full picture from last season. And these are the results….
The Numbers (and some nice bar charts if you’re into that sort of thing!)
Starting with the most complete data, Chart 1 below shows the weekly numbers of price rises and falls for all 37 relevant GWs in the 2015/16 season. Overall there were 636 price rises in the season, against a whopping 2,133 price drops – that’s 213.3m in total drops and a net reduction in value of the overall FPL roster of nearly £150m. That alone is a startling figure representing a ratio of falls to rises of 3.35/1.
It’s hardly groundbreaking that the table confirms a much higher level of price variability in the first quarter of the season. However, the scale of the imbalance is again striking. 30% of all price drops in the entire season occur in the 5 GWs from GW3 to GW7, with an average of 129 drops per GW in this period, which is more than double the full season average of 58.
Meanwhile however, in the same period, the average number of price rises, at 20, remains quite close to the season average of 17, with the highest in the period being GW with 26 rises. And in the first confirmation of the doubts around the international break, GW5 is the lowest of the bunch with only 15 rises.
This early high of 26 is matched or beaten by both subsequent international windows, however with GW9 being the peak of the season on 32, and GW13 with 26. The result of this imbalance is a ratio of drops to rises of 6.5/1, almost double the full season ratio given above.
Taking a look back to the data over a similar period for 2014/15 shows a pretty similar tale, as shown in Chart 2.
Noting that the international break took place in GW4 that season (the same as this year) the data covers up till 2 weeks after the international break GW, mirroring the period for 2015/16 somewhat. Despite not having data for the rest of the season, it is safe to assume these were most likely the highest weeks for drops in 2014/15, with an average of 117 per GW from GW3 to GW6.
It’s notable that GW2 has a significantly higher drop rate than GW2 the following year (50 versus 18), possibly due to the earlier International window. The ratio of rises to drops GW3-6 is 6.4/1, almost identical to GW3-7 2015/16. Regarding price rises, the international break GW4 has the second lowest number of rises of the 4 gameweeks considered.
So, what about playing the game of banking those 0.2m rises on wildcard? Time for more charts! Chart 3 and 4 show the numbers of multiple price rises (i.e. 0.2 or 0.3 rises by one player) in each GW for all of 2015/16 (in Chart 3) and the first 6 weeks of 2014/15 (in Chart 4).
Looking at the full season data, it confirms that multiple rises in a gameweek are indeed now quite rare. They predominantly occur in the first half of the season – 31 from GW2 to 19 and only 15 from GW20 to 38. But importantly no more than 4 multiple rises occurred in any single gameweek.
Looking in detail again at the key GW3 to 7 period, no more than 3 of these multiple rises occurred. In the early weeks of 2014/15 it’s a similar story, with a max of 4 multiple rises in GW3, but on 2 to 3 in the other weeks. It’s pretty clear the banking price rises on wildcard week plan is now a busted flush.
But what about price drops? Well, without overloading you with another chart, they are fairly sporadic. In the 2015/16 GW3-7 period there were 8 multiple drops in GW3 (which was also the biggest week of the season for drops with 183) however the numbers for the subsequent gameweeks were 1, 0, 1, 2. What was notable however was that the ‘unpopular’ players tended to consistently drop week after week in this period.
Possibly the more telling analysis therefore, is to look at cumulative changes in price over this key early period either side of the first international window. The table below gives a summary of this which is examined in more detail beneath.
The following are the key points from the above table.
2014/15 – GW3 to GW6 (International Break GW 4)
- There were 516 price drops versus just 84 price rises.
- 101 players dropped 0.2 or more over this period, adding up to a total loss of value of £22.3m (Note this is a week shorter period that the 2015/16 this considerably less players making the second drop)
- 20 Players dropped 0.3m or more, giving a total loss in value of £6.1m
- These included preseason fave Bojan who was the only 0.4 faller, while the 0.3s included the previous seasons FPL royalty Yaya Toure, as well as notables like Mario Ballotelli, Olivier Giroud, Philippe Coutinho and Rickie Lambert.
Meanwhile on the price rise front
- Only 18 players rose 0.2 or more over the same period, netting a total gain in value of just £5.7m
- £4.3 million of this came from just 11 players listed below who rose 0.3 or more.
- £3.1m came from just 7 players who rose between 0.4m and 0.6m
|Ángel Di María||0.4|
*Stevan Jovetic is the only anomaly here, having risen and fallen 3 times each (after an Aguero injury, a few Jovetic goals and then promptly getting injured himself).
2015/16 – GW3 to GW7 (International Break GW 5)
- There were 644 price drops versus just 99 price rises.
- 200 players dropped 0.2 or more over this period, adding up to a total loss of value of £47.4m
- 63 Players dropped 0.3m or more, giving a total loss in value of £20.0m
- The 9 toxic players listed below dropped 0.4m or 0.5m and sucked £3.4m out of the teams who held them.
- Only 23 players rose 0.2 or more over the same period, netting a total gain in value of just £7.4m
- £5.2 million of this came from just 12 players listed below who rose 0.3 or more.
- £3.4m came from just 6 players who rose between 0.4m and 0.8m (the latter of course being 2015/16 golden boy Riyad Mahrez)
|Gnegneri Yaya Touré||+0.5|
As this snapshot of the highest fluctuating weeks shows it’s a case of pretty widespread shedding of value, with the gains much more concentrated. And while you might say that those holding Franck Tabanou deserved everything they got; the big droppers list includes some very mainstream GW1 and early season picks, with the likes of Ivanovic, Kane, Milner, Henderson, Trippier and Bolasie among those who fell 0.3.
So, last season, if you were unfortunate to have other priorities for your free transfers and were stuck with even 3 or 4 of the 63 highest droppers, and failed to capitalise on some of the high flyers like Kolarov, Yaya, Wilson or Gomis, the relative effect on your team value could well be upwards of £2.0m.
1/ It’s pretty clearly confirmed in Charts 1 and 2 that there is a drastically higher level of price drops in the early weeks of the season under the parameters in operation over the past two seasons. More than 6 drops to every price rise which is more than double the drop/rise ratio of the season overall.
This is clearly therefore a risky time in terms of retaining value if not played wisely. A quick glance around teams of some of the less serious players in your leagues around, say GW10, and it would not be uncommon to find team values of 98m or even lower. Ignore the price drops at your peril!
2/ It also seems pretty clear that the days of banking profits from clever player trading on a wildcard are over, and have been over for two seasons. With a max of 4 multiple rises in any of the GWs considered in this analysis, and considering that 2 or even 3 of those are likely to be players you will want to retain in your final wildcard team, then the maximum likely gain on such ‘bandwagon riding’ is only 0.1 to 0.2m.
This is a fact that much of the ‘hardcore’ FPL world has yet to awaken to, with several reputable advice websites still outlining this practice as a good strategy and reason for an early wildcard.
3/ Was my suspicion of International Break wildcards justified? Yes, and no – and maybe. Clearly the first international break is generally woeful for price rises. Worst of the early weeks in 2015/16 and second worst in 2014/15, only 1 and 3 multiple rises in those seasons respectively. So yes. But the second and third international window gameweeks in 2015/16 were above average, and in fact 2 of the 3 highest of the year.
So no, although still far from a bonanza, and given that there are less drops to avoid, a wildcard is still more use in the first window. And, as I have no data pre 2014/15 I will never know the profile from those seasons or whether the new rules on wildcard transfers are the reason for the low rates in that first window.
So maybe! Of course there are many other valid reasons for wildcarding across the first international break and, if not motivated by the myth of a price rise bonanza, it’s still a good strategy.
4/ The team value game has changed. Despite the new landscape of price changes and limited number of rises, the choice of whether or not to wildcard in the highly price variable early season period can still have a significant impact on what budget you have to play with for the remainder of the season. But not in the way we had become used to over the years.
Rather than banking big gains from price rises, it’s now about timing your move right to avoid wiping out your value on the huge wave of dropping players, while surfing over a period of a few weeks (rather than just your wildcard week) on the tighter group of players who are hoovering up those gains.
As with many aspects of the FPL I don’t think there’s any hard and fast rule about when to wildcard. And if you have been wise/lucky enough to have picked a few of the early popular players, and avoided the worst of the value vampires you may not fare too badly value wise with those 4 or 5 FTs.
However there are other advantages of wildcarding early, perhaps this season more than most. Summer international late returners and an unusually high number of new managers at top clubs add to the normal uncertainty of unreliable preseason form, last minute signings and of course international break injuries.
But it’s worth considering that if you commit to and plan from the start for an early wildcard (rather than a vague intention to keep it as long as you can, before inevitably caving in GW3 as is my own pattern), then you can pick a team (based on fixtures, defensive rotations etc) tailored around the first 3 or 4 GWs and aim to maximise the points returns in that period.
And with the second wildcard no longer limited to January, and therefore temptation to cling onto the first one for DGWs, I can’t see an overall more advantageous time to play the first wildcard.
When exactly it is played will still depend on a number of factors, and in an ideal world the transfer window would be closed and everyone home safe from international duty, however if I find myself without this season’s flavours of the month, and with too many heading for drops I will not hesitate as early as GW3.
If you haven’t already, please check out the Football Survivor game – It’s free to play and there’s an FFPundits Invitation League on there for you guys to join and the winner will receive a Premier League shirt of their choice. It’s a fun game for us all to play in between the the FPL chaos. Follow the banner below to join our league or click here
Article written by @TrevG1977