Ten games in: the first proper review

Ten league games feels like the right point to stop reacting to individual results and look properly at what this version of San Diego actually is. I deliberately kept the same basic shape and instructions through the opening block because I wanted a clean sample to judge rather than constantly moving the goalposts. There have been game-state changes, of course, but the underlying structure has remained the same. The result is encouraging: we are sixth in the Western Conference, thirteenth in the Supporters’ Shield, with four wins, five draws and one defeat. The xG table has us around fourteenth on expected points, so we are slightly ahead of the underlying numbers, but not to an extent that makes the whole thing feel artificial.

Before the season started I set five fairly simple KPIs against what the club had produced in 2030, so the easiest place to begin is there.

The possession target is the obvious miss, but I am not sure I am particularly bothered by it anymore. We are only at 47%, yet sit sixth in MLS for pass completion at 90%, and we lose the ball less often than almost anybody. The Data Hub has us attempting only 503 passes per game compared with a league average of 563, so this does not look like a team incapable of retaining possession. It looks more like a team that is selective about when it has it. I would rather keep the quality of the passing and improve what happens after progression than start chasing an arbitrary 50% for the sake of ticking a box.

The attacking KPI is much more satisfying. Last season the team produced around 1.33 xG per game and scored only 0.82. After ten matches we are up to 1.48 xG and 1.30 goals per game, so both sides of that target have moved in the right direction. We are not an elite attacking side yet, but there is a very clear improvement in process and output. The slight frustration is that the conversion rate is still only 8%, twenty-fifth in MLS, which explains why some performances have felt better than the scoreline.

That fits what I have been seeing during matches. We can progress the ball, reach the final third and beat players. We are second in MLS for dribbles, first for fouls won, fifth for cross completion, top half for chances created and twelfth for final-third passes. The issue is not getting near the opposition goal. It is what happens once we arrive. There have still been too many sequences where the wing-back reaches the byline and the penalty area feels surprisingly empty. The next refinement needs to be about the relationships ahead of the ball rather than simply adding more possession.

The clearest individual discovery has been Tahiel Peralta on the left as the Wide Forward. I had wondered whether the eventual solution would be turning him into an Inside Forward, but the last few games have made me much less certain of that. His output across the ten games is already strong: 2.08 xA, 2.58 npxG, 2.82 key passes per 90 and 5.65 dribbles per 90. More importantly, he just looks more complete in that left-sided role than Rotondi has. He can receive, carry, combine and then make the final action himself. With his passing, vision, flair, first touch and movement, the role seems to bring together more of his strengths.

That does not mean Rotondi has suddenly become a problem. His overall production remains useful, and the Montréal game showed another possible use for him when he dropped into the central striker role against a back three. The point is more that Peralta may have answered part of the left-side question without requiring any structural change at all. If he can become a genuine source of penetration and creation on that side, the attack becomes much less dependent on what Culbreath produces on the right.

Culbreath himself has probably been the standout tactical success of the opening ten. His creative output from wing-back has been extraordinary: 4.19 xA, 0.42 xA per 90, 3.53 key passes per 90 and more than eight dribbles per 90. He is not merely holding the width. He has become one of the principal creators in the side. That is a huge positive, although it also tells me that one of the next jobs is creating a reliable second route to goal so shutting down Culbreath does not shut down too much of the attack.

I have also changed my mind slightly about Salvo. The idea of using him as a Deep-Lying Forward was attractive because of the empty-box issue, but his actual numbers suggest the opposite. He has generated 3.14 npxG, 0.41 per 90, and nearly four shots per 90. He is getting into positions to score. The problem is that he has only converted two of them. His movement and anticipation are stronger than his passing and vision, so dropping him away from goal might actually remove the part of his game that is currently working. For now I am more inclined to keep him central and work on the players around him.

The midfield is another area where I want refinement rather than reinvention. Müller has been too important to disturb. His progression numbers remain excellent and he is clearly the player that connects the first phase into the attack. What I do not want is to make both midfield roles so conservative that we solve one defensive problem by creating an attacking one. The better solution may be to keep Müller as the constant and alter the player beside him according to the opponent. Lohan offers running and stability; Vera gives me a very different option when we need more invention against a compact side.

Defensively, the headline number of eight conceded in ten matches is excellent, but the underlying picture is more mixed. We have allowed 12.55 xGA, so Paulsen and a little favourable variance have helped. We are conceding 12.3 shots per game, almost exactly on the original target, and only 54 shots on target in total. That is not a bad defensive platform at all, but it is probably not quite as dominant as the actual goals-against column makes it appear.

The interesting part of the Data Hub is that we make relatively few defensive actions, possession gains, interceptions and clearances compared with the league. That suggests the side is defending more through positioning and structure than constant ball-winning. I am comfortable with that as long as the chances conceded remain controlled. The main thing I want to study over the next block is whether the wide spaces that have looked vulnerable in individual matches are a recurring structural issue or simply the natural risk of asking the wing-backs to provide so much attacking width.

The one clear failure is set pieces. I set a target of at least three goals from corners, and after ten games we have scored none. Defensively the routines have been excellent, with no goals conceded from corners and above-average set-piece efficiency, but the attacking side has not delivered. That is not something I want to write off as bad luck given how much time went into the routines. The next block needs to tell me whether the issue is delivery, first contact, occupation of the six-yard box or simply the wrong players attacking the key zones.

So the ten-game conclusion is fairly simple: the system has earned the right to be refined rather than replaced. We are creating more than last season, scoring more than last season, passing the ball far better, conceding fewer goals and sitting comfortably in the playoff places. The problems are also becoming increasingly specific. To move the numbers again, there are five tactical changes I now need to consider:

1. Make Peralta the primary left-sided Wide Forward, with Rotondi used more selectively rather than assuming the left side belongs to him.
2. Keep Salvo central, but adjust the relationships around him so there is at least one more consistent runner arriving into the box when the wing-backs reach the byline.
3. Use the second midfield role more aggressively when the opponent allows it, rather than always pairing Müller with another conservative presence.
4. Build a second creative route away from Culbreath, so the right side is not carrying such a disproportionate share of the chance creation.
5. Rework the attacking corner routines, because zero goals from corners after ten matches is a clear miss against the original target.

For Games 11–20, I am going to tighten the benchmarks rather than rewrite them. I want to push possession towards 48% or better while keeping pass completion above 89%, maintain at least 1.45 xG per game, bring xGA below 1.15 per game, get shots conceded under 12, and improve the conversion rate from 8% towards 10% or better. I also want the team to remain top half for chances created and final-third passing, while watching much more closely who is actually arriving in the box when those attacks develop.

The first ten games were about finding out whether the ideas survived MLS. I think they have. The next ten are about making those same ideas sharper.

Two ideas for the next version

The ten-game review has left me with one fairly clear attacking problem. We are progressing the ball well enough, keeping it securely and reaching the final third, but too many attacks then slow down into sideways circulation. Eventually somebody gets bored, takes a shot from distance and the whole move rather fizzles out. I do not think the answer is to make the entire tactic more direct. The build-up is working. What I need is more incision once we arrive.

The first idea is to make the left side behave much more aggressively. Culbreath’s Advanced Wing-Back role has been one of the biggest successes of the season, so the thought is to mirror the width on the opposite side by using Bombino as an Advanced Wing-Back, with Lohan dropping into a proper DM role to replace the central security Bombino currently provides when he inverts. That would then allow Peralta to sit inside Bombino as a playmaking winger, receiving in the left half-space rather than being responsible for holding the width himself.

The attraction is fairly obvious. Bombino stretches the pitch outside, Peralta receives between the lines, Salvo stays ahead of him, and Lohan gives Müller the protection underneath. Instead of reaching the left side and immediately recycling, there should be a player receiving beyond the opposition midfield who can actually turn and hurt them. Peralta has the passing, vision, flair and technique for that job, while Bombino becomes the runner outside him. The whole idea only works, though, if Peralta is receiving with movement beyond him. If everybody still comes towards the ball, I have simply moved the sideways passing into a different part of the pitch.

The second idea is more aggressive again. Bombino remains the Advanced Wing-Back, Lohan stays at DM, but Peralta and Vera become two genuine 10s in the AMC positions. That would leave the two wing-backs responsible for the width, with Peralta and Vera occupying the half-spaces behind Salvo. Against a compact side, that gives us a proper front five across the pitch rather than asking one or both of the attacking midfielders to drift out and provide the width themselves.

The danger with that version is probably even clearer. Peralta and Vera cannot both become players who simply come short, receive, turn back and ask Müller for the ball again. One of them has to be the creator and the other has to be prepared to penetrate, run beyond Salvo and actually attack the box. Otherwise I will have built the most elaborate possible version of the exact problem I am trying to solve.

So the first idea feels more like a possible evolution of the base tactic: Bombino outside, Peralta inside, Lohan providing the security that allows it to happen. The second feels more like a weapon for games where the opposition have decided to sit in and make us break them down. Neither changes the basic principles. Both are attempts to make the final phase more decisive.

The main thing I am trying to get away from is possession for possession’s sake. We already know we can progress the ball. The next question is whether we can make the opponent move once we get there.

One of the things that has bothered me watching the first ten games is how vulnerable we can suddenly look once an opponent gets through the first line of pressure. The defensive record itself is good, but the underlying numbers are less dominant than eight goals conceded would suggest, and the more I look at the data, the more I think the issue may begin further forward rather than with the centre-backs themselves.

The current OOP setup is aggressive: High Press, Higher defensive line, Trigger Press More Often, Counter-Press and Step Up More. In theory that should squeeze the pitch, make the opponent uncomfortable and allow the back three to spend most of the game defending on the front foot. Bielsa-esque. Front foot defending. The problem is that the attacking players do not appear to be executing that press consistently enough for those instructions to work as intended.

Culbreath is the obvious exception. His pressures completed – 22%, possession-win and defensive-action numbers are all strong, and Peralta is reasonably active too, winning 20% of his. But elsewhere the picture is much less convincing. Salvo – 20% but barely involved in any overall defensive numbers, Ríos – 9% pressure success rate – and Rotondi – 28% but not exactly active either – are not producing particularly aggressive pressing numbers, and Vera – 29% win rate – is more active than effective in terms of actually turning pressure into possession wins. That leaves me with a situation where the team instructions are telling us to press like an elite front-foot side, but the personnel at the top end are not always capable of making that first action stick.

When the press works, everything behind it looks fine. When it does not, the problem becomes obvious very quickly. The opponent escapes the first pressure, the midfield screen is suddenly exposed, the defensive line is still high and looking to step forward, and the back three are asked to defend a much bigger amount of space than they should be. With the old box-to-box version of the midfield, that could effectively leave Müller and the three centre-backs trying to deal with the whole transition once the first press had been bypassed.

That also changes how I look at the centre-backs themselves. I still think there is room to improve the personnel, particularly in the middle of the three, but I am less convinced that simply signing a better defender solves everything. If the structure repeatedly asks the back line to defend after the first two layers have been beaten, even a very good centre-back is going to spend too much time dealing with ugly situations.

The ten-game data supports that concern. For all the aggressive pressing instructions, we are 29th in MLS for possession won, 28th for interceptions, and below average for defensive actions overall. We are not actually behaving like a high-regain side. That is probably the clearest mismatch in the current tactic: the instructions say aggressive press, but the outcome is a team that still does most of its defending much deeper.

So the next OOP refinement may be about pressing quality rather than pressing quantity. I am considering whether the Line of Engagement needs to come back slightly so the press can form properly before it jumps, whether Trigger Press should return to Standard rather than More Often, and whether Step Up More is asking too much of the back line when the player on the ball is not under enough pressure. I would rather have fewer coordinated jumps than lots of individual ones that open the pitch behind them. This feels vanilla but we need to go back to the starting point and see what is working and what isn’t.

The midfield change may help as well. If Lohan becomes a proper DM alongside Müller, there is suddenly a genuine second defensive layer behind the first press. Instead of the press being broken and the opposition immediately arriving at Müller and the back three, there is another player screening the centre and helping absorb the first transition. That gives the players ahead of them more freedom to press selectively rather than desperately.

The important thing is that I do not want to abandon the aggressive identity. I still want San Diego to defend on the front foot. I just want the aggression to come from the right moments, the right players and the right triggers rather than from every global instruction being pushed towards maximum intensity. At the moment the tactic looks more aggressive on the screen than it consistently does on the pitch. The next job is making those two things match.

A small addition to the attacking group, but one I quite like. Ben Broggio has joined on a free after his contract demands came in at a very manageable level, and the attraction is less about finding an immediate starter than adding another profile to the right side.

At 24, he still has a little room to grow and the underlying attributes are useful: good acceleration and pace, strong balance and agility, decent dribbling and technique, plus enough flair and composure to make him interesting when he receives in tight areas. He can also cover both flanks and all three attacking midfield positions, which gives me another option without having to carry a specialist for every role.

The important bit tactically is that I do not need him to become another creator on the right. Culbreath is already doing that. What I want from Broggio is something slightly different: receive inside the wing-back, carry towards goal, attack space and combine quickly rather than always coming short for the ball. In that sense he should complement Culbreath rather than duplicate him.

There are weaknesses, particularly out of possession. His tackling, positioning and concentration are not strong enough for me to expect him to lead the press, and that is worth keeping in mind given what the first ten games have taught me about our defensive structure. His OOP job will need to be simple and supported by the shape rather than asking him to become something he is not.

For the cost, though, it feels like a sensible bit of squad building. He arrives with Resolute personality, Determination 15, good physical tools and enough technical quality to give us another attacking route. No transfer fee, low wages, and a role that should be very clear from the start.

Not a headline signing. More of a useful piece for the next version of the attack.

The next thing I wanted to investigate was Bypass Press. Playing Through Press had been secure, but it was also giving the opposition time to get organised. By the time the pivot received and looked forward, the attacking movement had often already happened, as you can see below. We had five players across the final line, very little depth between midfield and attack, and the usual sequence became long forward pass, then any other forward passes unavailable, recycle, sideways pass, then eventually a long shot.

Bypassing the press was worth trying because the idea was not to change the attacking shape, but to reach it earlier. If the ball could get into the advanced players before the opposition block had settled, the same movements might become more dangerous. What quickly became clear, though, was that timing alone was not the whole issue. We could still reach the front five earlier and find them standing on broadly the same horizontal line.

That pushed me towards moving the two wide attacking midfielders into central AM positions. Peralta became the freer of the two, with a Free Role and permission to roam, while Vera was asked to provide more structure alongside him. The spacing immediately looked healthier because the pivots finally had receivers between themselves and the striker rather than having to play directly into a five-man front line.

I initially tried Vera as a Channel Midfielder. On paper it looked right because it combined creativity with movement into the channel, but in practice he was getting too high and arriving almost alongside the striker. That recreated the same problem in a different form. He is now an Attacking Midfielder with Less Forward Runs, while Peralta keeps the Free Role and Roam From Position. The intention is for Vera to become the connector and Peralta the more unpredictable player who can move wherever the space appears.

The early signs have been encouraging. The two AMs are giving the pivot more obvious vertical passes, and the attack has more depth rather than simply width. Peralta and Vera are not constantly occupying the same line as the striker, which means the ball can move through an extra layer before reaching the penalty area. It still needs a proper run of league games before I draw any conclusions, but visually it looks much closer to what I was trying to create.

One side-effect has been a huge amount of crossing. We completed only eight of 51 in the latest test, which looked alarming until I realised Culbreath was still carrying an old individual setup of Cross From Byline + Near Post. He had completed only 6% of his crosses in the previous game and around 21% overall, so that instruction was almost certainly contributing to both the volume and the predictability. Both crossing instructions are now back to Balanced while I watch what the wing-backs actually do in the new structure. The shape should naturally free them outside because the two AMs attract defenders centrally, but I do not want “free wing-back” to automatically mean “cross immediately”. The next step is to see whether they choose better deliveries once that restriction is removed.

Whilst I don’t want to change things all at once – sticking to my original plan of revisiting data every ten or so games – one of the smaller changes I wanted to test after the ten-game review was whether I had made the wide roles too aggressive, particularly now that one of the bigger changes was to add another one into the shape. Culbreath’s Advanced Wing-Back role has worked brilliantly, so the temptation was to mirror that on the left and ask Bombino to do the same thing. In practice, though, it often left both wing-backs racing into the final line too early and made the attack feel pre-programmed.

So Bombino has dropped from Advanced Wing-Back to Wing-Back. The idea is not to make him passive or bring him inside, but simply to have him arrive from a slightly deeper starting point. He still provides the outside option, still overlaps when the space is there, but he is not constantly trying to become one of the highest players on the pitch.

At the same time, I have removed the explicit Overlap instruction. The more I watched it, the more it felt like I was telling the players to do something their roles and positioning were already naturally producing. Even though the experiment last one game, I reverted to my original shape, with Peralta and Vera operating inside the wide defenders as a Playmaking Winger and Inverted Winger, there is already space for the wing-backs to go beyond them. Forcing the overlap seemed to be adding another layer of aggression to a movement that did not really need it.

The San Jose game was the first proper look at that version, and the movement felt more productive. Bombino was still wide and available, but he was holding back a little more, even when play was built on the other side of the pitch, rather than waiting on the final line. Culbreath remained the more aggressive outlet on the right, which preserves the asymmetry that has worked so well. The result was a shape that looked less rushed and gave the players more natural decisions in possession.

The crossing numbers were also encouraging. After some games where we were throwing an absurd number of balls into the box, the volume came down significantly and the completion improved. That may also have been helped by removing some old individual crossing instructions, but the wider point is that the wing-back no longer feels like the compulsory end point of every attack.

I am not trying to make Bombino into Culbreath on the opposite side anymore. Culbreath can remain the aggressive creator. Bombino can provide width, support and timing. If that relationship keeps working, the left side may be better for being different rather than symmetrical.


One of the things I have been trying to refine in the attacking structure is the role on the right. With Peralta increasingly looking like the main creator, Culbreath providing the aggressive width and Salvo occupying the centre-backs, I do not really want another AMR who behaves like a second playmaker. What I want is a dropping connector: somebody who comes short, gives the pivot an easy forward pass, combines with Culbreath, and then helps move the attack on rather than becoming another player standing on the last line. It will give me more variety and volume in the front five, instead of becoming the static line that I see when we are not quite clicking.

The recruitment search has therefore been built around that behaviour rather than simply looking for the best-rated right-sided attacker. I have been filtering for xA, open-play key passes and chances created, then checking pass volume, progressive passing, possession loss and whether the player naturally comes deep for the ball. The interesting part is that the data has produced some very different types of player. Yuri Leles looks the strongest overall creator, Will Alves is a very good receiver and carrier, William Hurtado has a tidy connective profile, while players like Tyler Goodrham and Xavier Simons bring useful qualities but also obvious compromises. Some are better creators, some are better carriers, and some fit the behaviour more naturally than the position.

The obvious comparison, though, is Ben Broggio, because he is already here. He is not the best pure passer in the group, with Passing 10, and he does not yet have the full trait package I want. But he has enough of the surrounding qualities to make the role plausible: First Touch 12, Dribbling 13, Technique 13, Composure 14, Anticipation 13, plus the acceleration and pace to receive short and move again. He is also extremely flexible across the attacking positions and may even have some usefulness as a striker, which gives him more value than a specialist-only signing.

The development plan for Broggio is therefore fairly clear. I want to add Comes Deep To Get Ball, Plays One-Twos and Likes Ball Played Into Feet and see whether those traits can turn him into the connector the shape needs. I may add Plays Short Simple Passes at a later date, too, mainly to try and mitigate his weakest attacking attribute, The lad has been here three weeks and has already managed to add one trait – Likes Ball Played Into Feet to his game: a rarity to do unless he has some super versatile behind the scenes attributes. If that continues in this way and the plan comes together, he does not need to become Peralta. He just needs to become the player who helps Peralta and Culbreath function better.

That is where the recruitment question becomes more interesting. There are players on the market who are probably better in isolation, but the gap between “better player” and “better fit for this very specific job” is not always large enough to justify a major fee or a DP slot. Leles, for example, is much more attractive to me as a possible Rotondi replacement and senior attacking piece than simply as a Broggio upgrade. At 24, with strong creative output and the ability to play AMR, AMC and striker, he could justify a bigger investment because he would replace more than one function.

For the pure AMR connector role, though, I am increasingly leaning towards developing what I already have. Broggio is young enough to improve, cheap enough not to create a cap problem, and versatile enough to retain value even if the experiment does not work exactly as planned. Unless the recruitment team finds someone who is clearly superior at the actual job rather than just statistically more impressive, it may make more sense to keep the money for the bigger upgrades elsewhere in the squad.