News, updates and analysis
To simplify pricing and based on your feedback, we're consolidating our subscriptions down to a single Pro tier. Effective July 1, 2026:
A two-tier setup added decision friction without delivering proportional value: most users on Basic eventually upgraded to Pro for the model access anyway, and the “am I missing something on the next tier?” question kept coming up in support tickets.
One clear plan, lower price, every model included — that's the shape that matches how you actually use the site.
Questions? Drop a ticket from Support. We'll keep this post updated with any clarifications.
— The cs2predict team
We've added a new model to CS2PREDICT: SURGE. Unlike ELO, rank, or even SQUAD (which averages player ratings), SURGE looks at how fast individual players are improving — capturing a "hot-hand" signal that no other model on the site sees.
For each starter on a team, SURGE computes the difference between their K-D differential over their last 3 matches versus matches 4-6 before. This is the player's recent trajectory — are they trending up or down?
Then SURGE takes the maximum across the team's five starters. A team with one player who's significantly improved his K-D in the last week is more dangerous than ELO suggests — and SURGE captures that.
We ran a grid search over 400 hypothesis variants on 10 years of historical HLTV data — varying window sizes (3/5/7/10/15 matches), features (rating, ADR, KAST, K-D differential), and aggregations (mean/median/max/min). The winner across all robustness checks: recent 3 vs prior 3, K-D differential, team max.
On a 1,973-match holdout of top-tier matches:
That 0.11 correlation is the key result. It means SURGE is providing genuinely new information that ELO doesn't see. When two independent signals agree, their combined prediction is significantly stronger.
ELO captures team-level Win/Loss outcomes — but it's slow to update and averages across the roster. SQUAD captures the level of player ratings but not the trend. SURGE captures the trend specifically of the best-improving player, which is what often drives upsets at the top of CS2 (one star catching fire can carry the team).
SURGE is available to Pro subscribers. Predictions will start showing up on upcoming matches within minutes of this post — the refresh daemon has already picked up the new model.
As live data accumulates, you'll see actual production accuracy on the model's page.
Based on your feedback and requests, we've reshuffled the model lineup across subscription tiers — to make more useful models available right away without a subscription, and to give BASIC subscribers noticeably more options to work with. Big thanks to everyone who wrote in via support and Discord — this update is built directly from your suggestions.
Even without a subscription, you now have two additional models with different decision logic, ready to try on upcoming matches.
Five additional models that previously required PRO are now open to BASIC subscribers too.
See the full lineup per tier on the Models leaderboard. Keep the feedback coming — it genuinely shapes how we develop the site.
We've added 2 new consensus models to CS2PREDICT: NEXUS and TRIBUNAL. Both were discovered through live data analysis — optimizing over real pre-match predictions with zero data leakage.
Unlike base models that compute predictions independently from raw team stats, consensus models act as filters on top of existing models. They only make a prediction when multiple models agree with high confidence. The result: fewer predictions, but significantly higher accuracy.
NEXUS — APEX + ORACLE both >65% same direction
The sharpest double confirmation on the site. NEXUS only activates when both APEX and ORACLE are strongly aligned — meaning two independent multi-factor models see the same outcome with high conviction. In backtesting on live top-tier matches: 100% accuracy on 12 predictions.
TRIBUNAL — OPEN + PRO + ORACLE all >55% same direction
A three-model consensus built around ORACLE as the anchor, confirmed by both the OPEN and PRO predictors. These three models use different feature sets — when all three agree, the signal is remarkably consistent. In backtesting on live top-tier matches: 100% accuracy on 11 predictions.
The 100% accuracy figures are from a small sample of top-tier matches — treat them as promising early signals, not guarantees. Both models will fire rarely by design: NEXUS requires two models at 65%+ confidence, TRIBUNAL requires three at 55%+. Expect 1–3 predictions per week on top-tier matches.
As live data accumulates over the coming weeks, you'll see real accuracy statistics on each model's page.
NEXUS and TRIBUNAL are available for Pro subscribers.
We've added 4 new prediction models to CS2PREDICT, bringing the total to 21 AI models. Each was discovered through automated optimization across 3,400+ feature combinations tested against 667 completed matches.
APEX — ELO + Form + Streak + Strength of Schedule
Balanced model optimized for the best accuracy-to-coverage ratio. Combines team rating with current momentum, win/loss streaks, and opponent difficulty. 66% accuracy, 60% coverage.
ORACLE — ELO + ELOHQ + Form + Tilt + SOS
Five-factor model that adds psychological pressure (tilt) and dual ELO systems. Performs best when teams have clear form differences. 69% accuracy, 51% coverage.
EAGLE — Rank + ELO + ELOHQ + Form + WinRate
High-confidence model that combines three ranking signals with team form. Predicts less frequently but with strong conviction. 77% accuracy, 14% coverage.
PHANTOM — ELO + TierELO + WinRate + Tilt
Maximum accuracy model. Only makes predictions when all signals align strongly. When PHANTOM speaks, it's almost always right. 92% accuracy, 6% coverage.
All four models are currently in calibration mode — they need at least 10 confident predictions before appearing in the model rankings. You can already see their predictions on individual match pages.
The new models are available exclusively for Pro subscribers.