Valorant Match Analysis: Map Pool, Rosters and Reading the Line

VALORANT looks like Counter-Strike from a distance — two teams of five, a bomb site, attack and defence halves, a first-to-thirteen map — and the analysis that works for one mostly works for the other. The differences are the interesting part. VALORANT’s competitive map pool rotates on the developer’s schedule, agent compositions change what a map is from one roster to the next, and the VCT calendar concentrates the top teams into a few international leagues and events. This guide works through a VALORANT series the way Sportslyx reads it: format, the market line where a public book prices the match, the map pool and veto, compositions and rosters, and finally how a read is graded and why a 1–2 is a push.

Format first

VCT matches are best-of-three; playoff finals are best-of-five, and some qualifier stages still use best-of-one. The format decides what a favourite is. In a best-of-three both teams get a map of their own and the series often turns on the decider; in a best-of-five a deep map pool is worth more than a single dominant map; in a best-of-one the veto leaves each side little control and the map itself can decide the favourite. Only a best-of-three or best-of-five can end with the loser one map short — 1–2 or 2–3 — and that scoreline is the basis of the grading rule at the end of this guide.

A map pool that rotates

The competitive pool is a fixed set of maps that the developer rotates several times a year: a map leaves, a new or reworked one enters, and every team’s pool changes at once. Two things follow. Results on a map that has just left the pool are irrelevant to the next event. And a map that has just entered has almost no history, so a team’s record on it is a few series at most and every roster is learning it in public.

Within the pool, a team’s depth is the number of maps it will play willingly. A team with two strong maps and a permanent ban has a predictable veto that opponents can plan around; a team with five playable maps can spend its bans attacking the opponent instead of protecting itself. The rotation also resets depth: a team whose strongest map rotated out is shallower than its recent results suggest, and that shows up in vetoes before it shows up in scores.

Reading the veto

The best-of-three veto follows a familiar pattern — each team bans one map, each picks one, each bans one more, and the remaining map is the decider. The picks tell you where each team expects to be favoured; the bans tell you what each fears; the decider is the map both were willing to leave and is usually the closest of the three. Between comparable teams the structure pushes series towards 2–1: each side wins its own pick and the decider settles it. If a favourite wins any one map 60% of the time and maps were independent, it would win the series about 65% of the time, lose 0–2 about 16% of the time and lose 1–2 about 19% of the time — one series in five ending with the favourite a single map short.

Compositions, sides and the shape of a map

What separates VALORANT analysis from Counter-Strike is that a map is not one thing. The agents a team chooses — the mix of duelists, initiators, controllers and sentinels — decide how a map is played, and a roster change often means a composition change on every map. A team that swapped its controller player over the break is running different smokes and different executes, and its old results on the map describe a plan it no longer uses.

• Side balance: most maps favour one side, and a team that wins its attack half on a defence-sided map is doing something its raw round total does not show
• Pistol rounds and their follow-ups: the first round of each half and the two after it decide a large share of maps, and a team that consistently converts them is winning structure, not variance
• Overtime: a map tied at twelve goes to overtime and is won by two rounds; a team with many overtime maps in its recent record has been in more coin flips than its map score suggests
• Compositions on the decider: the map neither team picked is the one where preparation is thinnest and default compositions matter most

Rosters and the off-season

VALORANT rosters turn over heavily between seasons and again at roster windows during the year, and the league structure means a single change can reshape a team’s whole map pool. Head-to-head results from before a change belong to a different team; results in the first events after one are small samples. The Sportslyx board reads rosters and team crests from bo3.gg’s public feed alongside the schedule, live map scores, results and the tournament and best-of format, so checking the current five against the ones behind the recent results takes a moment and removes most of the worst errors. Stand-ins, a coaching change and a new in-game leader are the changes that move results furthest.

Reading the series line

Where a public bookmaker covers the match, Sportslyx reads three markets from a free public feed: the series moneyline, a map handicap — usually ±1.5, so −1.5 needs a 2–0 and +1.5 covers with a single map — and a total-maps line, usually 2.5, which asks whether the decider is played. The moneyline becomes the win chance on the match page by a fixed calculation. Suppose a series is priced −160 and +135. The implied chances are 61.5% and 42.6%, summing to 104.1%; the 4.1 points above 100 is the bookmaker’s margin, and dividing each figure by the sum removes it: 59.1% for the favourite, 40.9% for the underdog.

That 59.1% is the market-implied chance shown on the public Valorant matches page. It is a statement about the market with its margin stripped out, not a Sportslyx pick, and it exists only where a book line exists. The free feed prices the top divisions; bo3.gg lists the VALORANT slate deeper than that, into tier-two leagues and qualifiers no public book quotes, and for those series the product works from a model estimate marked “est.” that is subscriber content and is not shown publicly. A public match page without a line shows no number rather than a synthetic one.

How Sportslyx grades a VALORANT read

• A read wins when the chosen team wins the series
• A series lost by exactly one map — 1–2 in a best-of-three, 2–3 in a best-of-five — is a push: returned, not counted as a loss
• A 0–2 or 1–3 is a loss; a best-of-one has no push, because a single map cannot be lost by one map
• A series without a final map score is voided rather than guessed

The push rule is more generous than any settlement a public book offers on a series moneyline, and it lifts the published hit rate relative to a plain win-or-lose count; Sportslyx states it openly rather than presenting the lifted figure as a straight record. Reads are selected by expected value with a price floor: the favourite’s calibrated series chance — calibrated because maps within a series are correlated and a raw series probability overstates a favourite — is set against the available price with the push treated as a returned stake, and a series qualifies only if the expected return clears a fixed threshold. A favourite priced too short is passed over, and best-of-one series rarely qualify because they carry no push.

Hit rate is wins divided by wins plus losses with pushes excluded. ROI is at a flat stake and labelled “partly est. prices”, because some priced series carried a model estimate rather than a book line. The record is published by calendar month in UTC on the Valorant predictions page and the Pick Tracker, and tracked picks of your own close fifteen minutes after the scheduled start.

A checklist before the series

1. Confirm the format: best-of-one, three or five
2. Check the current map pool and whether a rotation has just added or removed a map either team relied on
3. Read the market-implied chance where a line exists; treat an unpriced qualifier as unpriced
4. Compare both rosters with the ones that produced the recent results, and discount results from before a composition-changing move
5. Map the likely veto: permanent bans, picks and the probable decider
6. Weigh form by opponent quality and map score, and log any read of your own before the start

Summary

VALORANT analysis starts where Counter-Strike analysis does — format, line, veto — and then asks the questions the rotating pool and agent compositions add: which maps are current, which roster produced the record, and what the decider is likely to be. On Sportslyx the win chance is market-implied where a public book prices the series, the AI breakdown is written around it, and every board read is graded from the final map score with a one-map loss counted as a push. The Valorant analytics hub holds the current board, the methodology fixes the definitions, and the esports board runs the method on every series of the day. Past results do not predict future results, and no outcome is certain.