The short answerOn our ratings, Celtic and Beşiktaş are close to equal. Celtic are 91st in the world on the Beaumont Index and Beşiktaş are 109th, about a tenth of a goal per 90 minutes apart on a neutral pitch. The Europa League table tells a different story. Celtic have lost all three games and sit 35th of 36. Beşiktaş are unbeaten and eighth. Bookmakers make the visitors favourites at Celtic Park. Our model makes Celtic slight favourites at home, winning roughly 40 games in 100, with Beşiktaş winning about 32. Celtic's three defeats were a bad run against decent sides. Beşiktaş's seven points were a good run by a side that is not much better than Celtic.
That is not the mood around either club. Celtic were beaten 2-0 at Falkirk on Sunday, their sixth defeat in ten matches in all competitions. Supporters' groups put the board through a two-hour meeting afterwards, and Martin O'Neill's position is being questioned openly. Beşiktaş arrive having beaten Marseille 4-1 in Istanbul and won 3-2 at Hoffenheim, then drawn 2-2 with Crystal Palace. On Thursday, then, a club in crisis hosts a club on a run.
That picture is true, but it is not the whole of it. It is the part you see in results, and results over three European games carry very little information. The ratings use a lot more, and they say this game is much closer than the mood suggests.
Two clubs nearly level
The Index rates every club on one scale: goal difference per 90 minutes against an average club from Europe's top five leagues. Zero means a typical big-five side.
Celtic sit at about +0.01. On a neutral pitch, that is a club we would expect to match an average Premier League, LaLiga or Bundesliga side. Beşiktaş sit at about -0.11, a little below that average. The gap between them is about 0.12 goals per 90 minutes. That is small. It is about the gap between two mid-table sides in the same league.
Both ratings come with wide ranges. Celtic's plausible world rank runs from about 54th to about 155th. Beşiktaş's runs from about 63rd to about 194th. The two ranges overlap almost completely. On our numbers, either club could be the better one, and we lean slightly towards Celtic.
Then add the ground. Celtic Park is one of the louder venues in European football, and our ratings include a standard home advantage of roughly a third of a goal. Put a 0.12 edge together with home advantage and Celtic become slight favourites. Our model has Celtic winning about 40 games in 100, a draw in about 28, and Beşiktaş winning about 32. At the prices quoted this week, the market has those numbers the other way round.
The league each comes from
It is natural to assume that the Turkish champions-chasers must be stronger than the Scottish champions, because the Süper Lig is the stronger league. It is. Our league ratings put the Turkish Süper Lig around 15th in the world and the Scottish Premiership around 32nd. The gap between the leagues is close to four-tenths of a goal per 90 minutes.
But clubs are not their leagues. Celtic play most of their games against much weaker sides, and our ratings take that into account. What keeps Celtic's rating up is how much they dominate those games. Compared with the rest of the Scottish Premiership, Celtic are well over two standard deviations above average for deep completions (passes that reach the area near the opponent's goal), for pass volume and for touches in the opposition box. Few clubs anywhere control their league that completely. Once we discount the league, that dominance still adds up to a side about as good as an average big-five club.
Beşiktaş are third in the Süper Lig on our numbers, behind Galatasaray and Fenerbahçe. The table currently has them fourth. Their rating of -0.11 is what you would expect from a good Turkish side that is not one of the two best: dangerous, well stocked with experienced players, and below the level of the clubs that go deep into the Champions League.
So the usual instinct, a side from a better league against a side from a weaker one, works the wrong way here. The league gap is real. The club gap is not, because Celtic are much stronger relative to their league than Beşiktaş are to theirs.
What three games can and cannot tell you
Celtic lost 3-1 at home to Ferencváros, 4-1 away at Benfica and 2-1 at home to Celta Vigo. Nine goals conceded in three games is a real problem and nobody should dismiss it. Only one side in the competition has conceded more.
But look at who those games were against. Benfica are a strong Champions League-level side, and losing in Lisbon tells us very little. Celta Vigo are a LaLiga side, and a home defeat to them, after Celtic took the lead through Mika Baur and then conceded twice in three minutes, is close to what our model would expect from a game it rates as roughly even. Ferencváros at home is the result that hurts most. Celtic should usually win that game.
Read against the ratings, Celtic's European record is about three or four points worse than expected. That is a bad run, not proof of a bad team. Three games is not enough to separate the two.
Beşiktaş's run goes the other way. Beating Marseille 4-1 in Istanbul is a strong result. Marseille are rated well above Beşiktaş on our numbers, and nobody would have expected a three-goal margin. Winning 3-2 at Hoffenheim is a good away result, although our ratings have this Hoffenheim side slightly below Beşiktaş. Drawing 2-2 at home to Crystal Palace, rated about eight-tenths of a goal better, is a decent point. Seven points from those three games is about three more than our model would expect from a side of Beşiktaş's level. It is a good run, and that is mostly what it is.
When one team has taken three or four points too few and the other about three too many, the table can make two nearly equal sides look like a mismatch. Three games in, that is where this fixture stands.
Domestic form runs in the opposite direction from Europe
There is a second turn, and it matters because it cuts against the argument above.
Celtic's league points have run ahead of their chances for years. In our latest full model run, they had 24 Scottish Premiership points from nine games, from chances worth about 21. Last season they took 82 points from chances worth about 77. The season before, 92 from about 80. That is one of the most persistent surpluses we see anywhere. Some of it is quality the numbers do not fully capture, such as finishing and game control against weaker sides. Some of it is the advantage of playing a lot of games against teams that are happy to lose by a goal.
The recent results suggest that surplus is now fading. The 2-2 draw at home to Dundee United, where Celtic were booed off, and the 2-0 defeat at Falkirk both came after our latest full model run. If those results reflect a real drop, Celtic's rating will fall when we next refresh. A drop of a tenth of a goal or so would bring them level with Beşiktaş, rather than slightly ahead. Celtic are still top of the league, reportedly four points clear of Rangers, so the slump is real but it has not yet cost them the title race.
Beşiktaş's domestic numbers point the other way. In our latest run, they had 15 Süper Lig points from eight games, from chances worth about 17. That is a shortfall of two points, which is ordinary bad luck rather than a warning sign. Last season they were a couple of points ahead of their chances, and the season before almost exactly level. Beşiktaş are what they look like domestically: a solid, attacking side that has been slightly unlucky. Their away record is the weak point. They have won only four of nine away games in all competitions this season, and all three of their dropped results in their last five league games came on the road, including throwing away a 2-0 lead at Kasımpaşa on Sunday.
So Celtic's domestic results flatter them and their European results undersell them. Beşiktaş's domestic results undersell them a little and their European results flatter them. Put those together and you get two clubs much closer than the table suggests.
Where the game will be decided
Celtic's game is control. They want the ball, they move it into the final third and they get bodies into the box. That works against Scottish sides that defend deep. In Europe, the space it leaves behind gets punished. Most of the nine goals Celtic have conceded came from opponents breaking into that space or winning the ball back high up the pitch.
Beşiktaş are built to do exactly that. Under Vincenzo Italiano they press high and attack quickly, and they have scored nine goals in three Europa League games, joint-second most in the competition. Leandro Trossard is likely to play wide or behind the striker. Wilfried Ndidi gives them a ball-winner in midfield. Up front the choice is between Dušan Vlahović, their top league scorer, and Oh Hyeon-gyu, who has scored three of his four goals this season in Europe. Oh's history adds a little spice. He spent time at Celtic, mostly coming off the bench, and knows the ground well.
Beşiktaş also have gaps. Captain Orkun Kökçü and Fabio Miretti are injured, and centre-back Ümit Akdağ is suspended after his red card against Palace. Losing Kökçü takes away their best passer from deep. Losing Akdağ breaks up a back line that has already conceded five goals in three European games. A Beşiktaş side scoring three a game and conceding nearly two is not one that shuts games down. That suits Celtic, who have scored in eight of their nine home games this season.
Celtic are without Alistair Johnston and Yang Hyun-jun. Cameron Carter-Vickers' return at Falkirk matters most. He is the defender best able to deal with direct running behind, and Celtic's European record without a settled centre-back pairing has been poor. Expect O'Neill to change several players after Sunday, with Kasper Høgh, Sebastian Tounekti and Benjamin Nygren among those pushing to start.
The pattern of the game is fairly clear. Celtic will have more of the ball. Beşiktaş will try to break quickly and will have chances. The question is whether Celtic's defence, rebuilt again, can hold a line for 90 minutes. On their European evidence so far, that is the biggest risk to our model's lean.
What a result would tell us
The league phase has 36 clubs, and the top 24 go through to at least the knockout play-offs. Celtic on zero points after three games are not out. But a fourth defeat would leave them needing something close to a perfect last four games, and no side has yet done that from that position since this format began. For Celtic this is close to a must-win game.
For Beşiktaş the stakes are lower and the upside is higher. A win would put them on ten points and in the mix for a top-eight finish and a direct path to the last 16. A draw would still leave them well placed.
That difference matters. The team with the most pressure is also the team our model slightly favours. Nobody should be surprised if Celtic play nervously and Beşiktaş take advantage. Nobody should be surprised either if Celtic, at home, with their best defender back, finally play to their level.
What we think
Our view is simple and slightly against the grain. On our numbers, this is close to an even game with a small home edge. The table makes it look like a mismatch because three results each have gone the "wrong" way. Celtic have been unlucky in Europe and lucky at home. Beşiktaş have been lucky in Europe and slightly unlucky at home. Take out those runs and you have two similar clubs.
We give Celtic about a 40% chance of winning, a draw about 28% and Beşiktaş about 32%. If you think Celtic's recent slump is real, and there is a case for that, move a few points from Celtic to Beşiktaş and you get a true coin toss. It is hard to get from our numbers to Beşiktaş as clear favourites.
The more useful lesson goes beyond Thursday. Early in a 36-team league phase, the table mostly tells you who has had a good month. A side from a weaker league that controls that league as completely as Celtic do can be every bit as good as a good side from a stronger one. When the results and the ratings disagree this sharply after only three games, the ratings are usually closer to the truth.
Kick-off at Celtic Park is 20:00 UK time.
FAQ
Who are the favourites for Celtic v Beşiktaş?
Bookmakers make Beşiktaş favourites at Celtic Park. On the Beaumont Index the clubs are almost level, and with home advantage our model makes Celtic slight favourites (about 40% win, 28% draw, 32% Beşiktaş).
Where do Celtic and Beşiktaş rank?
Celtic are 91st in the world at about +0.01 goals per 90; Beşiktaş are 109th at about −0.11. The gap is about a tenth of a goal, and the 80% ranges overlap almost completely.
Why does the Europa League table look so one-sided?
Celtic have zero points from three games; Beşiktaş have seven and are unbeaten. Three games is a small sample: Celtic's European run is a few points worse than expected, Beşiktaş's a few better.
Is Celtic's domestic form better than their European form?
Yes on results: they still lead the Scottish Premiership, but their league points have long run ahead of their chances, and recent results after our last model run (Dundee United, Falkirk) are a warning.