Forrest was the first to double digit goals and assists for Celtic and leads Scoring Contribution with 27. Sinclair has 7 goals in five matches. Which of the wing wonders is having the best overall season?
Scoring Contribution and Expected Scoring Contribution
Both players have featured heavily with Sinclair completing the equivalent of 26.36 matches and Forrest 32.11.
In that time Forrest has racked up 11 goals and 16 assists.
For Sinclair it is 13 goals and 7 assists,
However, we don’t count penalties in the goals tally as that is a specialist skill that tells us nothing about finishing from open play. Sinclair’s goal tally therefore drops to 9 goals.
Overall Scoring Contributions per 90 mins are:
A clear actual contribution win to Forrest by 0.23 goals and assists per 90m.
But yet if we look at Expected Goals and Expected Assists (xSC combined) we get quite a different picture:
Remarkably we see almost the reverse pattern. Sinclair is expected to gain 0.12 more goals and assists per 90m compared to Forrest based on the quality of chances.
This all suggests Forrest has had a very hot streak that is likely not sustainable. Meanwhile, the flurry of productivity seen recently from Sinclair was expected and may continue. It isn’t predictable like that of course, but it is a fascinating result.
Goal Threat
Let’s dig underneath the pure productivity figures to see what is going on here.
Both players are scoring at exactly the same rate – 0.34 goals per 90m. As suggested above, Sinclair is under achieving versus xG (0.581 per 90m) whilst Forrest is scoring as expected (0.338).
If we look at their shooting statistics:
Across the board Sinclair is the more effective. He has more shots on target, and more taken inside the box (there may be a correlation!). His shots are more accurate and he coverts 4% more of all shots and 4% of shots on target.
By Sinclair’s underlying shot stats, a clear win for him.
Movement, Control and Positioning
As attackers, Celtic wide players need to be able to find space against deep and dense defensive cover. They need to be available for and control passes under pressure. Insight into the quality of movement, control and positioning can be garnered from a number of metrics.
Pack Receive is vital as this shows the number of forward passes the player brings under control and completes the next action. As attackers these players will often receive the ball in the opposition half and under defensive attention. Pack Receive is a good indicator of the quality of movement (are you available, can you find space?) and control (can you tame more difficult forward passes?).
In this regard Sinclair has the edge completing receipt of 8.12 forward passes per 90m compared to 7.82 by Forrest. The value of this (players taken out the game) is also higher for Sinclair scoring an average 41.81 Receive Impect per 90m compared to Forrest at 35.5.
Despite this, Forrest has possession of the ball inside the opposition box more frequently (5.23 per 90m). It is noticeable that Forrest is playing a more direct game this season and arriving late into the box when the ball is on the other wing. This is traditionally more Sinclair’s game and might explain Forrest’s improved goal tally.
Sinclair is more likely to lose the ball in the final third, and partly due to being twice as likely to miss control the ball than Forrest.
Also, Sinclair is over 5 times more likely to be caught offside.
Creativity
The wide attacker’s role is to create as well as score.
We have seen Forrest has an incredible 16 assists already. Sinclair was top of the assist charts with 15 for the season in 2017/18.
However, the same principles applies as for goals in that Expected Assists better reflects the quality of the chances created.
Another fascinating outcome as both players are achieving more assists than expected. Forrest is producing an assist every other game but by chance quality can be expected to produce one every three games.
Sinclair is also over achieving but less so. He is producing 0.3 assists per 90m compare to an expected value of 0.206.
Again, we can look at the fundamentals underneath this to see whether we think these trends will continue.
Forrest’s in game involvement is slightly higher and he completes around 3 passes per 90m more.
The Scot also completes very slightly more Pack Passes (forward passes that take out an opponent) than the Englishman with a resulting higher Pass Impect.
Forrest is creating 2.15 chances per 90m compared to Sinclair’s 1.29. Additionally, he is creating 1.59 secondary assists (the pass before the assister) to Sinclair’s 1.37.
No surprise then that Forrest provides 0.9 passes into the Danger Zone per 90m compared to 0.46.
Finally, completing a clean sweep of creative underlying stats, Forrest completes 0.37 crosses into the box per 90m compared to 0.19 for Sinclair.
Summary
The perception is that Forrest has had a great season but has perhaps gone cold whilst Sinclair is on the rise.
The reality is that Forrest is out scoring the Englishman but Sinclair can be expected to increase his goal return providing he keeps getting the same quality chances. Sinclair’s goal threat is consistently higher than Forrest’s.
Both players are over performing in providing assists. However, Forrest’s creativity exceeds that of Sinclair across the board although his assist rate will be difficult to maintain.
Expected goals and assists suggests Forrest will struggle to maintain his current rate whilst Sinclair may be due the goal rush he is on.
With Bayo, Burke and Weah in the building and Edouard to return it appears both have risen to the challenge of competition for attacking places based on Saturday’s performances against Hamilton Academical.
Either way, they complement the Celtic attack very nicely.
Graeme Mcmillan says
Well researched article. I am always impressed with the amount of work gone into collating the figures.
Are you able to do a correlation between xG and Actual goals, xA and actual assists etc, to see what the correlation coefficient is?
I have a suspicion that an analysis based on home games v away games could give a better insight given the relative pitch dimensions and the different defensive/attacking attitude of the teams involved.
Is Sinclair more effective home or away for instance?
celticbynumbers@btinternet.com says
Hi Graeme sorry missed the notification for this as was in my Spam folder for some reason!
Yes I track Goals and Assists versus Expected to see who is over and under performing. Browse the site there are various articles on this.
The bigger differential to numbers isn’t so much home vs away it is the standard of opposition especially Europe vs Scotland. I have in the past compared “Big Game” performance versus the rest where Big Game = Europe, Cup Semi and Finals and games versus top 6.
Graeme Mcmillan says
Hi Alan,
Thanks for the answer.
I am a numbers freak myself and use Python for data analysis.
Unfortunately I don’t have enough data for proper football analysis which is why I enjoy reading your excellent articles.
I take it that you don’t pay for data from opta etc. so that makes your work more impressive.
Duncan says
I think Sinclair is more of a goal hanger now than he was previously.
Forrest is more likely to hug the touch line and cut in creating a scoring chance than to be found hanging round the back post or 6 yard box like Sinclair seems to have developed this Season,
This for me comes down to confidence and ability.
Forrest is way more inclined to take men on around the outside and Sinclair less so as he often opts to play one two and cut inside to get a shot at goal.
This is not a criticism of either but I think of both adopted s bit of each others approach and added it to their own then their chances of scoring should increase.
Be good to see both flying at the same time as I don’t think many defences will cope with that in Scotland very well at all.
I also think the new faces and the emergence of Johnston have defo had a positive impact on Sinclair in particular.
Much like the new signing from Dortmund will have on Lustig and Izaguirre no doubt.
All good stuff again bud.