Ulsan HD — the most chances in the league, and the latest

00Why the table stopped me

This began with a single row

2nd — Ulsan HD · 26 matches · 12-4-10 · 40 points · 41 scored, 37 conceded · GD +4

Two things catch the eye. A second-placed side has lost ten matches. And its goal difference is +4. Seoul, first, are on +31; Jeonbuk, third, on +9 — a club below them with more than twice the margin. The points say second; the goals say fourth or fifth.

Usually that means one of two things: a side that keeps winning tight games, or one that loses badly when it loses. Ulsan are both. Several one-goal wins — and also two 1-4s, a 1-3 and a 2-3.

So the question set itself: how is this team second, and why is the goal difference only +4? What follows is the record of opening the data in order. The first two answers were wrong. I have left the wrong turns in. They were the most useful part.

01The first answer, and why it fails

Possession was measuring the scoreboard, not the style

Possession first. Ulsan average 43.6% in wins and 50.1% in defeats; the correlation between possession and points is −0.401. Season average 46.7%. Stop here and you get an answer: “Ulsan give the ball away and win on the counter.”

It is wrong. Change the window and it collapses.

02The second answer — half true at best

The same test, applied to “fewer entries, more shots”

Next, entries into the final third — completed passes that start behind the attacking third and end inside it. Ulsan make 26.6 per match; their opponents make 33.3. Yet Ulsan take 13.2 shots to their opponents’ 12.1. Fewer entries, more shots.

03Where Ulsan sit among the twelve

Season totals, which the scoreline moves less

04Shape of the attack — short and quick

Counted by possession spell

A spell ends when the opposition registers two consecutive events; that splits each match into possessions.

The last action before each shot tells the same story. 18% of Ulsan’s shots follow directly from a ball recovery (16% for opponents), while only 13% come from an open cross against 21%. Their entries into the final third also run more centrally — 25% through the middle against 18%.

05A different team after the break

The one pattern that survives controlling for game state

148 shots and 15.98 xG before the break; 194 shots and 27.00 xG after. Ulsan create 1.7× as much in the second half. The obvious objection is that they were chasing games.

Passing moves the same way: 16.9m average and 10.5% of passes over 30m in the first half, rising to 18.0m and 13.9% in the second. They lengthen — trading accuracy for territory.

06But they do not turn matches round

Strong after the break is not the same as a comeback side

In R2 they produced 2.47 xG in the second half alone and scored once. R15 is the opposite — 0.18 xG after the break, the only match where the second-half surge never came. Lead at the break and they win seven of eight; the exception is R23 against Gangwon, 2-1 turned into 2-3.

07Where the +4 comes from

Back to the opening question

If they create more than anyone, why is the goal difference only +4? xG attaches to each shot the number of goals it yields on average from that position and situation. Goals divided by xG equals 1.00 when a side scores exactly as expected.

08The empty column: set pieces

What 110 corners produced

09Creation runs through one player

Who produces the league’s best chance creation

Every main scorer finished close to expectation. Nobody is running hot. The 0.95 in section 07 is not one player’s poor month; it is the team’s average.

10Defending — most tackles, fewest fouls

Not about height, but about method

Two things arrive together. They win the ball cleanly — most tackles in the league, fewest fouls. And they do not press high — their defensive actions sit deeper than their opponents’, and they intervene in the attacking half and win the ball in the final third less often.

This is a side that sits and intercepts rather than one that hunts. And the ball it wins turns into a shot within fifteen seconds 55.9% of the time — which is section 04 again, from the other end.

The risk window: after 90 minutes

Six of the 37 goals conceded — 17% — have come after the 90th minute.

Five of the six came in matches they lost, so the table cannot separate “conceded while chasing” from “conceded through late lapses”. What it does establish is that the pattern repeats.

11Back to the table

The question was: how is this team second, and why is the goal difference only +4?

Second place is in sections 03 to 05. Ulsan do not circulate (eighth for passes) and reach the final pass in few touches (first for key passes). The ball they win becomes a shot within fifteen seconds 55.9% of the time, and half their shots end a spell of three passes or fewer. The result is the most shots on target in the league. Behind that, a defence that sits deep and takes the ball cleanly.

The +4 is in sections 06 and 07. That output is concentrated after the break, and from behind at half-time it does not convert into results (0 wins in 7). The chances they make return exactly what they are worth (0.95); the chances they concede return more (1.09), and at the edge of the box 1.45× more.

Ulsan create more than anyone in the league, and they create it later than anyone. Take a lead into half-time and they win seven of eight. Go in behind and no amount of second-half pressure has yet produced a win. Their ninety minutes are largely decided in the first forty-five.

And the first two answers in this piece were wrong. “They win with less of the ball” and “fewer entries, more shots” were both products of the scoreline. Recomputed over the goalless portion of each match, one reversed and the other mostly vanished. The first thing to doubt when reading data is not the data — it is the situation that produced it.

12Facing them, and their own to-do list

If you are preparing against Ulsan

1. The first half is the match. Ulsan’s first-half xG runs at 0.61 per game, 60% of their second-half figure. Get in front before the break and they enter the 0-from-7 pattern.
2. Manage the five seconds after you lose it. 55.9% of their shots come within fifteen seconds of a recovery and 47.4% within three passes. The danger is not their build-up; it is your turnover.
3. The middle. A quarter of their final-third entries come centrally against 18% for the sides they face. Blocking the half-spaces matters more than blocking crosses.
4. You can afford to give up corners. One goal from 110. No side in the league punishes them less.

Ulsan’s own list

1. The empty first half. The second-half output is already close to the ceiling. The points are in raising that 0.61.
2. 1.45× at the edge of the box. Variance or a real defensive problem — that needs separating.
3. 110 corners, one goal. The delivery and the finishing both exist; the link does not.
4. Reliance on Lee Dong-gyeong. One missed match is not enough to size it, but 23% of key passes and 27% of crosses is a risk on its own.

13Four matches that show it

Scored across seven axes, then narrowed to matches that carry the argument on their own

Possession and result are unrelated — won with 30%, lost with 60%. Entries and danger are unrelated — ten entries produced three times the xG of forty. One thing tracked the result: whether they led at half-time.

14A prediction instead of a conclusion

Four statements that can be wrong. The remaining twelve matches will mark them.

1. The goal difference improves. If 0.95 and 1.09 are variance, both drift back toward 1.0. As long as the chance creation holds, +4 is probably below what this side is. If the two numbers hold instead, it is structure, not luck.
2. The edge-of-the-box concession falls back. If it does not, there is a real problem in that zone.
3. A comeback arrives. Zero wins from seven is worse than 1.04 xG per second half deserves. If it stays at zero, the surge only works with a lead.
4. Extra points come from the first half. They will arrive in proportion to how far 0.61 moves toward 1.04.