We need to talk about Eamonn Brophy

courtesy of SNS

courtesy of SNS

written by: @TheGersReport

A quick look at the basic stat-line for Eamonn Brophy suggests that he is having a very good season.  He has scored nine goals (which is in the top ten) & averages 0.51 goals per 90.

However, rule #1 in fitba analytics is to strip away penalty goals from a player’s scoring rate.  This is not a new idea, but it still is one that folks are still getting used to…

Nine of James Tavernier’s eleven goals this season have been penalties

Nine of James Tavernier’s eleven goals this season have been penalties

Six years ago (!), Ted Knutson explained why it is imperative to remove penalties from any real analysis of a player’s goal scoring abilities. He wrote, “Use Non-Penalty Goals. This is the big one for forwards because it levels the playing field and focuses on what really matters.”

Each club likely has four guys on the pitch at any one time who can convert penalties at the standard rate, so why would you give any extra benefit to the forward who does the converting? If a defender is your penalty taker, and he scores 10 goals a season from the spot, but zero from anywhere else, does that make him good at scoring goals?

I understand some people care about penalty takers in terms of having a leadership or mental strength characteristic, and that’s fine. You can just tick a box on the player evaluation form that says “Takes Penalties” and be done with it. In terms of statistical analysis, you really want to filter out the penalty goals from the much harder earned goals that come via the rest of the game.
— Ted Knutson, Statsbomb (June 2013)

Remember how Eamonn Brophy averages 0.51 goals per 90 minutes…well, he averages 0.29 non-penalty goals.

He has nine goals this season & four of them have been penalties. 

tenor.gif

No…that is definitely not good.

The visual below will give you a sense of what 0.29 goals per 90 for a forward looks like in context. It shows the Goals Above Average per 90 for each forward who has played at least 1,000 minutes this season. These are players that teams have leaned on heavily to lead the front line.

Screen Shot 2019-04-29 at 6.59.33 PM.png

Goals Above Average (GAA) measures how many more goals per 90 a player scores above the average rate for a forward in that league.  The average goals per 90 for Scottish Premiership forwards since the 2013-14 season is 0.325 -  Eamonn Brophy has averaged 0.285 non-penalty goals per 90 which translates to a GAA of -0.04.  His production has been that of a below average forward.

For the record…the most ‘average’ forward in recent seasons was Rangers’ Joe Garner back in 2016-17 - his GAA was +0.009. 

That’s the definition of an average forward in the Scottish Premiership.

His Goals Above Average for Wigan in the English Championship is +0.03…Joe Garner screams “AVERAGE!”

His Goals Above Average for Wigan in the English Championship is +0.03…Joe Garner screams “AVERAGE!”

I first introduced Goals Above Average back in the summer of 2017 as a means of evaluating the signing of Alfredo Morelos.  We had read about his goal scoring record in Finland….but let’s face it…not many of us really follow that league closely & who knows…maybe people score goals for fun in the Veikkausliiga. 

His Goals Above Average as a 20 year old was +0.70 (nearly double his current rate of +0.37). 

Morelos was beyond elite for Finnish football & since then he’s shown that he is probably too good for the Scottish Premiership as well.  Recently players like Moussa Dembélé, Liam Boyce, & Louis Moult have proven to be too good for Scottish football & their top GAA rates were +0.29 for Dembele & +0.22 for both Boyce & Moult.  Last season, Morelos’ rate was +0.17 & this season it’s +0.37.

Of course, Eamonn Brophy is nowhere near those rates yet still has been linked to a move down south as recently as this past winter transfer window.  That’s largely based on his age & a pretty good season last year &….four penalties.

As Kilmarnock prepare for the summer, they need to decide whether to keep investing minutes in Brophy or to look to move the 23-year old while his market value is still relatively high. 

A stat like Goals Above Replacement (GAR) may help with that decision.

In that same Morleos blog post, I introduced Goals Above Replacement.  It’s not as easily digestible as Goals Above Average…but could be much more useful for clubs as they make decisions in the summer about their forwards.

The concept of being above average is pretty easy to understand.  But what about Goals Above Replacement?  

From a statistical point of view, a replacement level player is outproduced by 80% of forwards in the league.  Basically, you could slot in any forward from the reserves or even from a lower league & expect him to produce at this level.

Dawson Sprigings, who used to write about hockey analytics under the handle of @DTMAboutHeart & is now a consultant for the Colorado Avalanche (one of the surprise teams in this year’s NHL playoffs)explained why he prefers using the replacement level in his player analysis.  “I chose to baseline my statistic to ‘above replacement’ instead of ‘above average’ because it provides a quick way to decipher if a player is fit for an NHL roster spot.”  

If a forward is barely scoring above a replacement level player – do you really want him in the lineup?  Last season, Stevie May’s output was that of a replacement level player (his GAR was +0.02) & his underlying numbers suggested that there would be little to no improvement this season.

Well, given that he has not scored a non-penalty goal this season…his Goals Above Replacement is -0.20.

Another example of a forward with a negative GAR last season was Eduardo Herrera. 

Need I say more?

imgID141714044.jpg.gallery.jpg

Back to Brophy…let’s see where his numbers fall in the Goals Above Replacement universe this season.

Screen Shot 2019-04-29 at 7.24.08 PM.png

A replacement level forward averages 0.20 goals per 90 & Brophy averages 0.29…so his GAR is +0.09. 

There are forwards above that clearly need replaced (Rakish Bingham, Curtis Main, & Christopher Kane) & there are some that are said to be such a handful that goals scoring can be overlooked (Tony Watt, Dolly Menga, Uche Ikpeauzu). 

Then there’s that group who are slightly above being replacement level forwards - Florian Kamberi, Danny Mullen &……Eamonn Brophy. 

Kamberi was due for a regression in scoring, which was probably made much worse by the human relations disaster that was Neil Lennon with Hibs this season.

I don’t know if it’s a confidence thing or a lack of application, but Flo’s nowhere near as hungry as he was last season. He needs to find that hunger again because he’s an important player for us
— Neil Lennon, December 2018 (Edinburgh Evening News)

Someone may want to slip Modern Fitba into Lennon’s DMs next time.  The numbers pointed out that Kamberi’s scoring rate wasn’t going to be repeated but also suggested that he would still be a good forward for this level.  Of course, numbers can’t influence a player’s reaction to constant negativity from his “leader” & his numbers plummeted this season.

Danny Mullen has made a steady career progression from League One & the Championship with Livingston & then St. Mirren (despite basically being an average forward at the Championship level - his GAA ranges from +0.03 in 2014-15 to -0.07 in 2017-18).

Then there’s Eamonn Brophy, a player who had a breakthrough last season with Kilmarnock & a Goals Above Average of +0.13 (basically on par with Louis Moult in 2016-17 & Anthony Stokes in 2014-15).

_105555562_18343503.jpg

So, who is the real Eamonn Brophy? We should start with looking at his career goal rates & then should analyze the underlying numbers to see what they tell us about his prospects moving forward.

Below you’ll find the annual Goals Above Replacement & Goals Above Average rates for Eamonn Brophy…from his dominant spell as an 18-year old in League Two on loan with Queen’s Park, to his years with Hamilton & then to his breakout with Killie last season.

Screen Shot 2019-04-29 at 7.11.42 PM.png

Brophy’s goal scoring rate in League Two was the best for any forward since 2013-14…which quickly flagged him as being ready for top flight football. However, outside of last season, the youngster has yet to truly produce as a top forward in the Premiership (or even the Championship during his brief spell with Dumbarton).

Given that Brophy is still so young & is now only entering the prime years of his career…the question becomes what his true ceiling as a goal scoring forward will be. When you look at this season in isolation…there is some real concern.

The visual below shows where Brophy ranks with the other forwards in the Premiership who have played at least 700 minutes.

Screen Shot 2019-04-29 at 6.25.39 PM.png

This really is a bit of a clusterfuck.  His goal rate is actually above average for forwards (this season) & his Projected Goals (explained here) is at an elite level for Scottish Premiership forwards.  But then…everything else is basically in the red - signifying numbers that are either below average or at a replaceable level.

giphy.gif

I’m as confused as those of you who actually stuck around this long to keep reading this. 

Basically, it looks like Brophy’s goal rate (as “meh” as it is) is largely driven purely by shot volume.  He averages 3.08 shots per 90 (4th amongst forwards with 700+ minutes) & averages 0.74 Scoring Chances per 90 minutes (which is the 12th best rate).

That’s the beginning of our disconnect. 

Brophy gets on the end of a lot of shots…but there may also be a lot of noise there.  Only 24% of his shots are high percentage chances (which ranks 31st!!!!!!!!! among forwards). 

Basically, Brophy has a shot selection issue.

Screen Shot 2019-04-29 at 6.33.48 PM.png

When I reviewed footage of Brophy’s shots this season…beyond the obvious ‘hope & a prayer’ shots from outside the box, I noticed a lot of shots that were inside the box - but at a bad angle - like the one above.  Brophy does get this shot past the defender…but given where he’s taking it — he’s gonna need a wonder goal to beat the keeper.

This helps explain why ALL of Brophy’s finishing rates above are so very, very bad.

  • Brophy has scored on 7% of his shots, his Expected Conversion Rate is only 12% (both are bad for a forward at this level).

  • His Shot Accuracy is beyond bad…20% of his non-penalty, non-free kick shots are on target.  20%!!!  His Expected Shot Accuracy (based on the kinds of shots he takes) is 42%…which is the 27th best rate for forwards (& I use the term ‘best’ very loosely here).

  • When Brophy does manage to get shots on target…his Shooting Percentage (the rate of shots on target that are goals) is 29% which is basically aligned with his Expected Shooting Percentage (xSh%) of 31%.  His xSh% is the 28th best rate for Scottish Premiership forwards.  Twenty……..eighth…….

This number caught me the most off guard.  When Brophy does get a shot on target to test the keeper…his expected success rate is really, really low.  This made me look back to the rate from last season…when he was among the best forwards in the league.

Well…

In 2017-18, Brophy actually only scored on 28% of his shots on target which basically aligned perfectly with his xSh% of 27%. 

Basically, what this all means is something that Modern Fitba colleague, Matt Rhein, pointed out in his preview post touting Brophy as a ready-made replacement for an aging Kris Boyd:

There are certainly things that Brophy still can work on. When we look at Eamonn Brophy’s Expected Shooting Percentage last season, mentioned on this site previously, we see it was 27%. He finished with an actual shooting percentage of 28% so we would not expect any type of serious regression based on this, but these figures are a bit lower than we would like.

This is likely due to Brophy seeming to fancy a long range shot. Comparing him to the rest of the top 20 in xG per 90 minutes in the SPFL this season, we see him 10th in the percentage of shots he has taken in the danger zone (or the area in the 18 yard box in between the 6 yard box where you are more likely to score) at 48%. At 5 feet 9 inches, Brophy cannot rely on his size to force his way into quality shooting positions. He will need to figure out a way to get better shots in the future, though that could come with more experience on the pitch.
— Matt Rhein, Modern Fitba (June, 2018)

The scoring regression actually did come…once Brophy began missing the target more - his poor shot selection came more to the fore. Maybe, given his lack of size, he was unable to live off of creating chances from more dangerous areas & when the randomness of football caught up to Brophy…we found a striker hoping to score from low percentage areas on the pitch.

I decided to stroll over to Matt Rhein’s cubicle in the Modern Fitba offices to see what he thinks & he replied by whipping up Brophy’s shot map this season.

PHOTO-2019-05-01-07-47-52.jpg
This past summer, I had written on this web site that Eamonn Brophy was the perfect player to fill the Kris Boyd role at Kilmarnock. This season Killie seemed to concur, giving 1,459 minutes in league play through 35 matches where he only appeared in 1,113 minutes all of the 2017-18 season.

However, even with the increase in playing time, Brophy’s goal scoring out width penalties has slowed down with a non-penalty goals per 90 of 0.29 compared to 0.45 last season.

Brophy is averaging a similar number of shots per 90 minutes as he did last season, but the type of shot he is taking has changed. In 2017-18, Brophy averaged 0.15 xG per shot but this season he is averaging 0.12.

Last season, 48% of the Killie striker’s shots came in the area of the pitch known as the danger zone (where goals are more likely to be scored) while this season only 35% of his shots are coming from that same area.

For some reason, Eamonn Brophy is not taking the same number of shots from dangerous areas as he did this season. Are defenses locking in on him to prevent those chances? Is he “letting it fly” after a few long range goals more?

Whatever the cause, the effect is less goals. Brophy has shown he can get these dangerous shots before. If can he can do it again, he can see those goal numbers increase again.
— Matt Rhein, May 2019
courtesy of SNS

courtesy of SNS

Ultimately, I remain hopeful that Brophy is a player that can be among the better forwards in Scotland. His agility & quickness gives him a skill-set that should be able to exploit defenses who are still anchored by slower center-backs who retain their roles due to their aerial dominance in a cross-heavy league. Smaller, quicker forwards could thrive as they play at a speed that forces defenders to play on their heels while trying to defend the likes of Brophy.

The red flags mentioned above regarding his underlying numbers should trigger a process of performance analysis that can lead to coachable moments - both in the video room & on the pitch. Why is it the shots he does get on target lead to such a pitiful Expected Shooting Percentage? Shot selection is definitely part of it…but why isn’t Brophy getting more of his higher percentage shots on target?

This is the kind of inquiry that data can trigger within a club to help young players, like Brophy, meet his apparent potential.

Some notes…

  • Stats are courtesy of data provided by Ortec Sports.

  • Stats don’t include this past weekend’s matches.

  • Florian Kamberi is another one whose numbers are being dragged down by his inability to regularly test the keeper this season (that & the fact that he is only averaging 1.97 shots per 90 - which is in the bottom half of forwards in the league).

Given the types of shots he takes, Kamberi should really be testing the keeper more. Last season his Expected Shooting Percentage was 36% (which he outproduced at such a rate that regression was obvious) but he also struggled to get shots on target last season (he had a shot accuracy of 38%). Kamberi tends to finish his shots on target at a higher rate than expected…the next step, like Brophy, is looking for ways to get more of his chances on target.

Given the types of shots he takes, Kamberi should really be testing the keeper more. Last season his Expected Shooting Percentage was 36% (which he outproduced at such a rate that regression was obvious) but he also struggled to get shots on target last season (he had a shot accuracy of 38%). Kamberi tends to finish his shots on target at a higher rate than expected…the next step, like Brophy, is looking for ways to get more of his chances on target.

  • I just noticed that the link to Dawson Sprigings’ blog post has been removed from the world wide web…that often happens when analysts get hired by teams (can’t give away the secrets). I’ll leave the link up just in case it ever gets republished.

  • Here’s a visual aid to see how many forwards have scored at the different Goals Above Average rates over the past five full seasons. It can help give you a sense of what’s elite, what’s very good, & what’s really bad when looking at the GAA rates in this post.

Screen Shot 2019-04-30 at 1.05.51 PM.png
  • I got called out online a little bit for suggesting that Kyle Lafferty would suck this season & despite the fact that he kind of has (just ask his manager)…his actual goal rate is surprisingly high. His Goals Above Average, albeit in limited minutes is actually, +0.15 - which you see in the visual above is really good.

Screen Shot 2019-04-30 at 1.10.05 PM.png
Screen Shot 2019-04-30 at 1.10.25 PM.png
Screen Shot 2019-04-30 at 1.07.24 PM.png
Screen Shot 2019-04-30 at 1.07.32 PM.png
  • This was written under the influence of the new Fat White Family album, CSNY, & KARP.