Bitcoin Dota 2 betting sites

Dota pricing is meta-first

Every title prices team-vs-team; Dota 2 additionally prices the meta. Over a hundred viable heroes, balance patches that reshape the viability map, and team metas that take days to weeks to settle - books' models trail that settle cycle, and the lag is the Dota calendar's most reliable feature. 20 books on this page carry documented Dota 2 markets, and the patch cycle below is the frame that organizes both their menus and their mistakes.

The calendar: TI, the EWC window, and the between-seasons

1. The International (August). Valve's world championship - the largest

single-event Dota book of the year. TI 2026 ran in Shanghai (Team Spirit's third title, 3-2 over Team Vision) [1]; the prize-pool era has contracted deliberately from the 2021 crowdfund peak (US$40M) to a leaner 2026 pool (US$3.36M) [1] - handle relevance now rests on the event, not the prize number.

1. Esports World Cup (July-August). The Dota club championship inside

the festival (the Riyadh Masters line) - the summer futures market. EWC 2026 ran in Paris after the announced venue move [2], with ESL supervising formats [2].

1. Between-events. ESL One-class events and regional play carry the

calendar; tier-2 regional cards appear mainly at esports-first books [7]. The TI event page and EWC page track the cycle; Valve's official esports hub [5] carries the tournament schedule.

The menu, tier by tier

Live values from our database - details and caveats in each review. "-" = not documented by the operator. "Visit" links may be affiliated.

Baseline menu everywhere: match winner, map winner, map handicap, map totals [8]. The class separators:

• Draft-phase props - first blood, first tower, kill totals, correct

map score. The esports-first verticals carry them on tier-one matches [6] [9]; sportsbook-class lists stop at winner-plus-totals [8].

• Live per objective. 95% of this page's books run live

Dota. Roshan takes and item timings are the repricing anchors; kill markets react first, objectives lag. The live hub carries the cycle mechanics.

• Outright spread. TI outrights at every class; EWC/ESL One futures at

most; regional futures only at the esports-first verticals [7].

~6.5% is the current median sampled margin across this page's books; 5.8% the sharpest documented sample [7] [8].

The patch-cycle edge, concretely

The cycle repeats two to three times per tournament season [3]:

1. Patch drop. Hero reworks and talent changes reset the viability map.

Team metas are unknowable for days - prices during this window carry the year's widest model uncertainty.

1. Meta settles. Competitive play reveals the working hero pools over

days-to-weeks. Books' models adjust with a lag - hero-pool picks and map-total drifts are stale first here.

1. Stable window. Prices sharpen again. The edge flips: fade community

overreactions, not the model.

Two practical notes sharpen the cycle read. First, major patches near big events compress the cycle - teams arrive with settled metas, and the stale window is shorter. Second, hero-rework patches move map totals more than match winners: pacing changes (ganks, objective speed) show up in totals before they show up in series prices.

The EWC window and the summer market

The Esports World Cup's Dota club championship is the summer futures market between TI editions, and it prices like its own event class rather than a TI warm-up:

• Club-format quirks. The EWC club structure (points, club rosters

spanning titles) creates team-selection questions the regular circuit does not have - squads that optimize for the club championship behave differently from squads optimizing TI. The outright market prices this unevenly.

• The venue-move variable. EWC 2026 moved from Riyadh to Paris on

short notice [2] - travel and prep-cycle changes that the market absorbed late. Event-logistics news is a real Dota input in the EWC window.

• ESL-supervised formats. Tournament formats are ESL-run [2]; the

format documents (group structures, tie-breakers) are the settlement rules by another name - read them before betting the group-stage markets.

The TI-to-EWC axis defines the Dota year's two big liquidity windows; everything else is the settle-and-thin cycle the patch section describes. The EWC event page tracks the festival window year-round.

The draft read, and why books underprice it

Draft-phase information is the Dota specialist's edge: pool signatures (teams have hero-pool identities), counter-pick patterns, and the supports-vs-cores drafting order all move win probability in measurable steps. Sportsbook-class Dota has no draft market at all, and esports-first verticals price draft props conservatively - the stepwise probability moves between pick screens are where the stale side survives longest. The same logic makes hero-pool news (nixed players, new signatures) a series-price factor that books handle unevenly.

Roshan and the live anchors

Dota live betting has one anchor no other title has: the Roshan timer. The aegis timing is public information on the game clock, and every live market reprices around it:

• Aegis-still-up states. Between Roshan's death and the aegis use or

expiry, team-fight win probability shifts measurably - the live series market is where that state is cheapest to trade.

• Roshan-pit setups. The 5th/6th Roshan windows (with cheese and

refresh) concentrate the late-game variance - books that suspend on Roshan contest states price them honestly; the rest carry a void clause as the correction.

• Between-map resets. Drafting for map 2 is a new game economically -

the between-map window is where between-map momentum mispricing shows first and disappears fastest.

The kill-score mirage is stronger in Dota than in LoL: net-worth and item timings carry the win probability, and the kill feed is the noisiest of the visible signals. The live totals market is the cheapest place to trade item-timing reads where menus allow it - which is again an esports-first verticals condition.

Warnings specific to Dota

• Patch-window temptation. The stale-price season is real, and it is

also when the least information exists. Bet the meta-settle window, not the patch-drop chaos - model uncertainty is symmetric.

• TI group-stage format quirks. TI's format has historically featured

long group stages with tie-breaker matches; void and settlement rules on abandoned/tie-breaker fixtures vary per book. Reviews quote the terms where operators state them.

• Regional depth drops off fast. Outside tier-one regions and the

flagship events, Dota coverage thins to a handful of esports-first books with wide margins - same specialist caveats as every tier-2 card.

• Bonus market restrictions. Draft props and specials are the exact

market class some rollover terms exclude. The bonus hub documents which terms restrict what.

How the books model Dota - and where they miss

The model class a book runs shows in specific market behaviors:

• Odds-feed copiers (sportsbook-class lists) publish tier-one Dota at

sane margins but thin everything below: no draft props, no tier-2, wide in-play spreads. Fine for occasional TI betting; structurally uninteresting for specialists.

• Vertical-first models price the meta cycle explicitly - their

patch-window behavior is the tell. A book whose Dota lines moved on the last major patch's second week (not the first) has a meta-aware model; a book that repriced instantly has a team-form model wearing esports clothing.

• Volume books (the widest menus) carry tier-2 and minor regions with

the wide margins those deserve - their tier-one prices are competitive, their specialty is coverage.

The three classes coexist on this page because they serve different bettors; the menu table above is the honest comparison surface. Sample your own fixtures - the market-margin tool recomputes any captured market on the spot, and captured samples beat class labels every time.

Where to go next

~6 esports titles per book is the current median across this page's operators; the books table filters on Dota 2 coverage live, and the odds guide carries the margin anatomy behind every sample on this page.