written by – Rangers Report
Back in 2014, 11tegen11 advocated for utilizing Expected Goals as a primary means of using statistics to identify potential goal scorers for teams to scout.
He wrote that, “there are two elements that constitute a good striker. (First), the striker has to get into good scoring positions, and accumulate good shots. This is best measured as Expected Goals per 90 minutes, with exclusion of penalties. [Secondly], the striker has to convert these chances into goals. This can be measured by comparing Expected Goals and actual non penalty goals.”
The idea of applying Expected Goals, rather than actual goals, to scout for goal scorers comes down to the fact that players’ ability to finish, i.e. their shooting percentages, has proven to fluctuate from season to season. Devin Pleuler, who now does analytics for Toronto FC, proved this back in a 2014 OptaPro Blog post, while Ben Pugsley, the co-creator of Statsbomb, made the same conclusion in a 2013 article. Pugsley demonstrated that a player’s ability to generate shots is far more repeatable then how many goals they score or at what rate they scored them.
Pugsley wrote, “There is virtually zero relationship between a striker managing to convert his shots on target into goals in year one to year two.”
Knowing that, let’s breakdown the top goal scorers from the 2015-16 Scottish Championship season – applying the same outline that 11tegen11 used in his How to scout goal scoring talent? blog post.
First, here’s the traditional goal scoring leaderboard (excluding penalties) alongside their shooting percentage (goals divided by shots on target).
All stats in this post are from league matches only.
|Mark Stewart||Raith Rovers||12||0.545|
|Derek Lyle||Queen of the South||10||0.394|
|Iain Russell||Queen of the South||10||0.579|
|Lawrence Shankland||St. Mirren||10||0.385|
|Blair Alston||St. Mirren||10||0.455|
|Stephen Mallan||St. Mirren||9||0.333|
|James Craigen||Raith Rovers||6||0.364|
|Calum Gallagher||St. Mirren||6||0.250|
|Gary Oliver||Queen of the South||6||0.400|
|Ryan Hardie||Raith Rovers||6||0.333|
Now the astute observer knows that comparing Martyn Waghorn to anybody this season is misleading given that he missed eleven matches due to injury. Plus, what about Ryan Hardie? He only featured in ten league games this season. You can’t group his six goals with someone like Will Vaulks who played nearly 2,400 more minutes.
That’s why it only makes sense to break down goal scorers based on their per 90 minutes averages. How many goals do they score on average, i.e. goals per 90 minutes.
To narrow down the list, I put the minimum number of minutes played at 700. (which meant Harry Forrester, who averaged 0.66 goals per 90, did not make the cut).
Some surprising names leap out. Christian Nade was unstoppable in his cameo for Dumbarton this season – scoring seven goals in 12 matches. He’s also 31-years old. Ryan Hardie, again, was one of the main catalysts for Raith Rovers superb second half of the season. He scored six goals in ten matches & remarkably 78% of his total shots were on target. Then there’s Bob McHugh scored at the same rate as Martyn Waghorn? Wait, what?
Can McHugh match that Martyn Waghorn’s production if given more time? This is where it is imperative to remember that a player’s ability to finish off his chances has proven to fluctuate from season-to-season. McHugh scored some high-profile goals for Falkirk this season & his goal-scoring rate is impressive. But he also saw 40% of his shots on target beat the keeper. Christian Nade’s shooting percentage was 0.636(!!), while Ryan Hardie’s was only 0.333. Which of these three are likely to see their scoring rates regress?
Even a player who played more minutes & still scored at a high rate, like Mark Stewart, is unlikely to sustain his scoring rate next season (0.545 sh%).
This becomes our entry point to factor in Expected Goals as a way to determine which players deserve a closer examination of their talents. The total Expected Goals leaders can be found here, but I think that we have already proven the importance of utilizing per 90 minutes stats when looking to evaluate performances.
*Mitch Megginson’s totals are from Raith & Alloa combined, while Tom Walsh’s are from Rangers & Dumbarton combined.
The top three of Waghorn, Anthony Stokes & Jason Cummings represent the three dominant forwards in the Scottish Championship last season. There are few that would debate that fact. Kenny Miller isn’t that far behind, while Dominique Malonga just couldn’t do enough to earn constant minutes for Alan Stubbs. The stats suggest he was one of their most dangerous players in the first half of the season, but ultimately he couldn’t get enough minutes to prove that worth (he averaged 62.5 minutes per game). He has since moved to Italy to ply his trade.
We should keep in mind something that 11tegen11 pointed out in his analysis, “This metric stands for creating goal scoring opportunities, which is a combined effort of both the striker himself and his team.” It’s no surprise to see 15 players on this list from the top three teams in the league – Rangers, Hibs & Falkirk. These sets of players clearly fed off of each other & the teammates around them. To a lesser extent Ryan Hardie & Mark Stewart likely thrived off of the second half surge that was Raith Rovers, but it’s also more likely that they were catalysts behind the winning form – particularly Hardie.
Players like Jordan White, Declan McManus, Jordyn Sheerin, Christian Nade, & Denny Johnstone were creating chances for teams that were routinely outshot & losing the possession battle. Should we factor that in when comparing them to players like Kenny Miller & John Baird?
Ultimately projections only go so far until players need to produce actual results.
Below you’ll find the difference between the actual non-penalty goals scored by each player & their Expected Goal totals. In other words, which players were scoring more goals then the projections suggest that they would, & which players were underperforming given the amount of shots (also the type of shots) they were getting.
Only players from the original xG/90 list are included.
Kenny Miller & Christian Nade were the best finishers in the Championship last season but you would assume that their play will regress going forward. Miller’s shooting percentage was 0.560, while Nade’s was 0.636. The average success rate of all shots on target in the Championship last season was 0.305. Nade is looking for a new employer next season & Miller will be moving up to the Premiership. These changes, & the fact that both are over 30, would seem to make it a safe bet that the duo to see their goal scoring form dry up to an extent going forward.
Mitch Megginson split his season between barely featuring for Raith Rovers & being the #1 striker for a woeful Alloa Athletic side. Throughout the season, the former Aberdeen youngster was able to generate quality scoring chances. Unfortunately, his shooting percentage was 0.143 – which seems more woeful than unlucky. The data suggests the goals are due to come…or he just isn’t good enough to compete at this level.
Wait, why are we talking about Mitch friggin’ Megginson? Don’t you see Martyn Waghorn at the bottom of this list, along with the coach of Hearts U20 team? Don’t be alarmed….yet. Waghorn took 107 shots this past season – which led the league (despite missing the last few months). Half of those shots were scoring chances (High Danger/Very High Danger Shots). That volume meant he had a high number of Expected Goals. Despite that he was tied with Bob McHugh in goals per 90 minutes – this was largely due to McHugh finishing off 60% of his scoring chances, while Waghorn only put 35% of his High/Very High Dangers shots past the goalie. Waghorn’s ability to create large numbers of scoring chances is likely to continue next season, however if his shooting percentage remains at this level – questions may be asked…
This was basically the longest introduction ever to narrow down a list of players that teams should consider scouting if they are looking to add potential goal scorers. Now, we all know that clubs like Rangers, Celtic, Hearts, Aberdeen, etc. probably won’t be looking to the Scottish Championship for goals.
Jason Cummings, who has a year left on his contract, may be the exception.
Given his high rate of Expected Goals per 90, his propensity for the ‘big goal’ (he scored the first goal in eleven games this season, the next best player did it seven times), & the idea that his play is still improving makes him an intriguing target. He scored under his Expected Goals totals, but did have a shooting percentage of 0.419 on scoring chances. His propensity to take long range shots (37% of his shots came outside of the goal area) dragged down his overall success rate. However, he was deadly when taking a shot (with his foot) in the scoring chance areas.
That kind of profile screams out for a move up to the Premiership…& since Hibs can’t offer that – the time may have come for the 20-year old to make the switch. Given his contract situation, it may be smart business for the Scottish Cup Champions to cash in before they lose him for nothing.
Besides Cummings, the top clubs in Scotland will be better served looking elsewhere for goals. But, the bottom six clubs & other teams in the Championship could find some gems off of this list. For the Premiership clubs, they could take a chance on a younger player to see if they can make that next step & score goals in a higher level. For Championship teams, there are players on this list who may not play regularly – but probably should & could thrive if given more minutes.
Obviously, the transfer season has begun & some players are already off the market. James Craigen has already signed with Falkirk, Blair Alston has joined St. Johnstone, 33-year old Iain Russell has signed with Airdrie, Jordan White has gone down south to Wrexham, & White’s former Livi teammate Jordyn Sheerin has signed with League Two side – Brechin City. Both White & Sheerin were in the top twelve in xG per 90 minutes, for a since relegated club, so the indicators are there that they are poised to thrive with their new employers.
Others are heading to the Premiership already – either via promotion (Rangers) or returning to their home clubs after time out on loan. For example, Anthony Stokes looks to be primed for a free transfer – but his wages may mean most of his suitors come from England. Meanwhile, Lawrence Shankland looks poised to get a chance to prove his worth with Aberdeen after a season on loan in the Championship.
So after excluding players like that from the list, the pickings were kinda slim.
But, the following players statistics suggest that they should be heavily scouted by teams on the look for (cheap) goals.
The 24-year old isn’t likely to be leaving Hibs this summer, but who knows what his role will be if Neil Lennon wants to bring in some of his own transfers to the club – especially if they lose Cummings. A bottom six Premiership club would be wise to take a flier on Keatings. Plus, the former Hearts & Hamilton forward really doesn’t have much else to prove at the Championship level. Over the past three seasons, he has scored 31 goals & has averaged 0.57 goals per 90 minutes. Last season’s numbers are inflated by his precision from long range, but the fact that he averages nearly three shots per 90 minutes & was 12th in the league in Expected Goals per 90 suggests he could easily be a scoring threat at the next level.
The 23-year old striker was brought in last summer by former Queen of the South manager James Fowler. Hilson was coming off back-to-back productive seasons in League One with Forfar Athletic, where he scored a combined 22 goals in 5,003 minutes, which averages out to 0.40 goals per 90 minutes. In his first season in the Championship he averaged 0.38 goals per 90. He didn’t necessarily generate a high number of shots in his limited time but the majority (56%) were scoring chances. He has a year left on his contract & wouldn’t cost much (valued at £59,000 on TransferMarkt), which makes him a low-risk transfer target for a Championship club that may be on the hunt for goals. Given that the new manager, Gavin Skelton, made signing a striker his first piece of business that seems to indicate that Hilson is there for the taking. Hilson’s shooting percentage was 0.333 (which is slightly better than average) & still was able to outscore his Expected Goals totals by 1.64 goals. It’s a small sample but there may be upside there….for cheap.
The winger is the right price: free. The 20-year old winger has seen his contract expire & will be looking to leverage his 2016 success with Raith Rovers into an opportunity to get extended playing time somewhere else. He joined on a free transfer from Dundee United in January & was part of the influx of talent that helped lead Raith’s surge to the promotion play-offs. Although he’s not a striker, he proved that he could complement the threat of an effective forward (Ryan Hardie) & pick his spots in the penalty box. His goal scoring rate of 0.40 goals per 90 minutes may not sustain, given his high shooting percentage (50%) but his ability to create chances should also factor into his evaluation. Given his age, his price tag, & previous Premiership experience – Connolly would be a smart investment for any team in the Scottish Championship. Of course, his former boss at Raith may be the first in line…
First off, I am a Ryan Hardie believer & this is not a plea for Rangers to sell off the 19-year old forward. However, he more than proved his value at the Championship level & there doesn’t seem to be a spot for him, yet, in the Rangers first team. So, the next logical step would be to send him on loan to another Premiership club to see if he can compete at the top level. The talent is there, but given the pressure that will be on Mark Warburton to produce instant success – he may not be ready to adapt a ‘trial by fire’ approach with the lanky teenager. However, if he can go out on loan for a few months & produce at a Premiership level – he will be too difficult for Warburton to ignore.
You can follow Rangers Report on Twitter @TheGersReport