Team Liquid and siuhy part ways: what 185 maps of data say about a failed leadership project

In a small apartment in Da Nang, I keep a notebook that is not used for scores. It records the things I refuse to bet on.

The first line was written in the summer of 2026, in pencil, and has not been crossed out: never bet on a team that is still searching for a leader.

On August 13, 2026, Team Liquid confirmed that Kamil "siuhy" Szkaradek was leaving the roster and entering free agency. A short announcement, few words, no long farewell. Behind it sit sixteen months, 185 professional maps, three Majors, and one number that forces anyone who reads data for a living to stop: 0.87.

That is siuhy's average HLTV rating across his entire Liquid stint. For an ordinary player, 0.87 is a warning sign. For a tier-1 In-game Leader, it is a verdict written before the match begins.

The more telling fact, however, lies elsewhere. Liquid has changed leaders four times in four years and still has not found the right one.

The IGL carousel: four names, one vacancy

To understand why a roster-move report matters, place it on a timeline.

In April 2026, siuhy joined Team Liquid on loan from MOUZ. The deal was notable in several ways. First, siuhy is Polish, trained in the European system that produces most of CS2's top tactical leaders. Second, MOUZ was one of the best-coached and best-structured organisations in Europe across 2026 and 2026. Sending an IGL from MOUZ to North America runs against history: European teams used to pull talent from North America, not the other way around.

Third, and most importantly, it was a loan, not a buyout. That structure signals commitment. A team that buys a leader builds its entire system around him for years. A team that borrows a leader is experimenting, and experiments can be cancelled at any time.

Before siuhy, the leadership role belonged to Casper "cadiaN" Møller and Russel "Twistzz" Van Dulken. After siuhy, from July 2026, it belonged to Johnny "JT" Theodosiou. Four names, four different calling philosophies, in one short cycle.

In any team sport, constant change at the helm points to a deeper problem. A football club that sacks three managers in two seasons is rarely firing three bad managers. It is firing because the club's structure will not let anyone succeed. Basketball calls this the coaching carousel. CS2 calls it the IGL carousel.

Dust2.us, in the original report, described Liquid as having several strong pieces around siuhy. That phrasing is worth recording. It is the only data point in the article that favours the player, and I will return to it later.

At the same time, the regional picture deserves attention. 100 Thieves was cited rebuilding its CS roster. s1mple was cited moving to BC.Game. Liquid itself, in Dota 2, had a vacant mid position after midOne's departure. Four separate events inside one time window are a sign of a broad roster-reconstruction wave, not an isolated case.

One more detail about the tournament system is needed to judge the scale of the failure. The Major is the world-championship tier event organised and sanctioned by Valve, gathering the strongest teams through regional qualifiers. For an organisation like Liquid, reaching Major playoffs is not only a competitive target. It is a condition for brand standing, sponsor persuasion, and player retention. A Major playoff run delivers viewership, content, and brand exposure that no third-party event replaces. Three Majors without playoffs means three missed commercial cycles.

The data does not lie, but it does not say everything

Now to the core.

siuhy's 0.87 is calculated across 185 maps. Emphasise the 185: this is not a small sample. A player with 185 professional maps has faced every kind of opponent, map, and pressure. A large sample removes luck. A low rating over twenty maps can be form. Over 185 maps, it is nature.

But this is where data analysis demands discipline. HLTV rating is a composite of kills, assists, and survival. For an IGL, it carries a role discount: the caller plays more utility, enters sites first to open space for teammates, and accepts deaths in exchange for information. In modern CS2, the IGL role has shifted toward the fragging leader who calls and kills. The discount still exists.

So what is the acceptable threshold? At tier-1, top teams generally tolerate an IGL rating between 0.95 and 1.05. Below 0.90 is a warning zone. At 0.87, even after adding the role discount, siuhy sits at the bottom of the range a tier-1 team can endure.

This does not mean siuhy was the sole cause. It means siuhy was not the answer.

The team record reinforces that conclusion. Across three Majors, Liquid reached the playoff stage zero times. At the most important CS2 events, the record was 5 wins and 11 losses. The best result was a top-four finish at FISSURE Playground #2, a third-party mid-tier event.

Sixteen months, three Majors, no playoff run. In sports statistics, sixteen months is long enough to rule out variance. Three independent failures at three separate events are not bad luck. They are a pattern.

One technical detail prevents a rushed conclusion. Majors use a Swiss group stage, largely BO1. BO1 amplifies variance: a single map can be decided by one shot, one misread, one random situation. In theory, a 5-11 record could be a format artifact.

But that argument collapses next to the playoff drought. If the problem were only BO1 variance, the team would have cleared the group stage at least once in three attempts. Playoffs are BO3, where skill and system matter more than luck. Failing to reach that stage even once shows the team could not peak under maximum pressure.

Structurally, CS2 differs from titles like League of Legends or DOTA 2. CS2 runs on physical mechanics: bullet paths, recoil, movement speed, buy economy. Valve updates the game infrequently and with small changes. There is no meta-reset mechanism comparable to MOBA balance patches.

The consequence: if Liquid failed for tactical reasons, mid-round calling, map-pool depth, or team coordination, the cause does not lie in a patch. It lies in people and systems. No update rescues a team that calls the wrong strategy.

Throughout my tracking, I apply a checklist before drawing any conclusion: which data supports the hypothesis, which data contradicts it, and which data is missing. Here the supporting data is clear: a low rating over a large sample, no Major playoffs, a losing record at big events. Only one data point contradicts: the claim that Liquid had strong pieces around siuhy. Much data is missing: no utility numbers, no opening-duel win rate, no mid-round calling quality.

That gap matters. It is why I conclude siuhy was not the solution rather than siuhy was the cause.

One memory returns whenever I analyse a case like this. Amid the roars of Russia, I heard a number whisper, and it was more accurate than the crowd. In the summer of 2026, aged fifteen, I watched France beat Croatia 4-2 in the World Cup final. The world said Croatia deserved it. But when I dug into the data, Croatia won three of six knockout games while their expected goals exceeded their opponents in all six. The team branded lucky was the team creating more.

That lesson followed me into esports. A team that wins through individual moments does not necessarily have a good system. A team that loses while controlling the match does not necessarily have a bad one. And an IGL with a low rating does not necessarily call badly, nor necessarily call well.

In 2026, when the pandemic closed stadiums, I collected data from 312 matches across six European leagues. Home win rate fell from 46 percent to 38 percent. The PPDA index, the passes allowed before pressing, rose by an average of 1.8 for home teams, meaning they pressed less without a crowd. PPDA is a lens, and through it I saw Morocco in the semi-finals two months early. I used that same approach to predict Morocco's 2026 World Cup semi-final run, while everyone laughed at my model.

The common thread: I was not searching for answers, I was searching for ignored variables. In the siuhy case, the ignored variable is organisational structure, not the individual player.

In 2026, interning at a small sports-data firm in Ho Chi Minh City, I wrote a twelve-page report on Spain's wide pair at the Euros, Lamine Yamal and Nico Williams. The data showed the pair generating 4.2 expected goals per match from inward dribbles, higher than any midfield pairing. Yamal received the ball 11.3 times per match when opponents pushed high, opening space for the full-back to overlap. That report went to three European betting firms, and a week later I received a part-time offer from a company in Malta. I accepted, but kept studying, because I believe systems built slowly last longer than systems built in haste.

That belief applies directly here. A team building slowly does not change leaders four times in four years.

In football, the only thing worth trusting is what the crowd has not yet seen. But experience also taught me the opposite: sometimes the crowd is right and the analyst is wrong for loving his model too much. My first big bet did not come from courage. It came from the crowd's mistake. That was the Morocco-to-beat-Belgium bet at odds of 5.80. I won. But I have also lost similar bets by confusing correlation with causation.

That is why the next section exists.

The contrarian angle: four pieces and one vacancy

The claim that Liquid had several strong pieces around siuhy is the most thought-provoking data point in the original report. If true, it partially reverses the story.

Picture two scenarios. Scenario one: siuhy called badly, the good pieces could not function, and the team failed because of its leader. Scenario two: the pieces looked good on paper but did not fit their roles, the organisation was unstable, and any leader would have failed. Both scenarios produce the same result, no Major playoffs, but their meaning is entirely different.

The available data cannot separate them. This is where I remind myself that my model is not truth. I have been right against the crowd many times, and those wins make it easy to form a contrarian bias, always wanting the surprising conclusion. Before writing any counter-argument, I ask: is this really true, or do I just want it to be?

With that question, the honest conclusion is this: four leaders in four years is the strongest evidence for scenario two. When an organisation changes leaders repeatedly, the probability that all four were poor is low. The probability that the structure is the problem is high.

This leads to another view of the move. siuhy's departure does not solve Liquid's problem. It only changes who is responsible for it.

At the same time, the finances and contract mechanics deserve a look. siuhy arrived on loan and left as a free agent. Liquid recouped no transfer fee. In accounting terms, this is an unrecovered investment, a failed loan rather than a realised buyout. On the other side, releasing an underperforming salary is positive cash flow for the club. On pure cost, the move is rational. On assets, it is a loss.

The loan mechanism leaves one unanswered question. If MOUZ retained siuhy's rights throughout the loan, his free agency could result from an expiry clause or a mutual termination. The report does not say. If there was an option-to-buy that Liquid declined, this is a commercial decision, not a rule event. And the absence of any reported transfer fee suggests the negotiating balance favoured the club over the player's leverage.

But the open roster spot is the real cost. At the time of the report, Liquid had not announced an official replacement. JT holds the role, but his promotion after siuhy's benching looks reactive rather than the product of a planned search. A vacant IGL slot gets more expensive the longer it stays open: the roster practises without a stable system, and every week that passes is a week of coordination not built.

This is where regional analysis matters. Liquid is the flagship organisation of North American CS. That it had to borrow a Polish IGL from MOUZ, and still failed to reach Major playoffs, signals that North America cannot supply tier-1 leaders from within. Historically, talent flowed from North America to Europe. Now it flows the other way. That is structural, not incidental.

If siuhy returns to Europe as a free agent, it becomes a repatriation flow. Europe recovers a Polish leader that North America could not keep.

One further note on reputational risk for both sides. For Liquid, a sixteen-month story ending in free agency is a negative mark on its brand narrative of stability. For siuhy, a 0.87 rating over 185 maps is the kind of signal that compresses a free agent's leverage, especially for an IGL whose value is hard to measure in raw stats. But the MOUZ pedigree retains its value. If he lands at a European project that needs a system-builder but cannot afford an IGL who both calls and frags well, this could be a value signing. In the transfer market, a low valuation often creates opportunity, not only risk.

What to watch in the next cycle

Three signals will decide how this story is retold.

First, who becomes Liquid's next IGL. If it is a proven big name, the story is rewritten as an upgrade. If JT continues, the story is a stopgap. If it is an untested internal name, the story is a rebuild. In all three cases, the real measure is not the name but the time it takes the system to run smoothly.

Second, where siuhy lands. With his MOUZ pedigree and the European system, he will most likely find a tier-1 or tier-2 project in Europe. This will be the test of the central hypothesis of this piece: if he succeeds in a new environment with a stable system, the correct conclusion is that Liquid has a structural problem. If he keeps sinking, the correct conclusion is that his individual level has declined.

Third, the reconstruction wave. Liquid, 100 Thieves, BC.Game, three moves in one window. If the trend continues, it shows organisations shrinking budgets and optimising rosters rather than expanding. For anyone tracking the transfer market, that signal is worth more than any individual rumour.

My pencil notebook still holds its first line. But perhaps I will add a new one, in ink: a team that changes leaders four times is not looking for a person. It is looking for a reason.