Style. Blitz specialist (5+0 time control, all games). Averages roughly 10 games per day sustained across nine months — high volume for an intermediate club-level player. Impulsive tendency: 40–48% of moves are played in under 2 seconds. Average per-move decision time is ~6 seconds. Wins tend to run 20–28 moves; losses tend to run 26–30 moves — indicating fight-through behavior rather than quick concessions.
Skill. Win rate is essentially flat at ~52% across the entire 9 months. Elo has cycled between 900 and 1500 during that same window, meaning the rating swings are being driven by behavioral variance, not skill change. A cold reader would conclude: this player's tactical/positional strength is stable — what moves the rating is state, not knowledge.
Volume patterns. Almost no zero-game days across 321 tracked days — very high consistency of engagement. Marathon sessions (20+ games in a single day) were common in the first three months, then essentially eliminated in months four through nine. Peak day was 33 games. Current typical day: 5–10.
Time of day. ~30% of games are played between midnight and 6am UTC — late-night activity is a persistent feature of the schedule.
Behavioral trajectory. Session-level volatility (StdDev) dropped ~55% over 9 months. Variance dropped ~90%. Rating cycles compressed — a pattern that once required 200 games now completes in ~80. Player is demonstrably self-correcting: after Elo crashes, volume drops sharply, then rebuilds.
The distinctive thing. The cyclical pattern of ascent → peak → correction is unusually consistent. Each cycle looks structurally like the last, just faster and smaller in amplitude. Rating gains and losses do not correlate with skill improvement — they correlate with behavioral state. This is someone whose chess rating is functioning as a real-time mirror of internal regulation, not tactical knowledge.
What This Shows
Your End-of-Day Elo rating plotted over time. This is the primary outcome line — everything else on this dashboard exists to explain why this line moves.
How to Use
Toggle the behavioral overlays below to layer metrics on top of the Elo line. Look for metrics that move BEFORE the Elo changes — those are your leading indicators (the drivers).
Key Findings From Your Data
StdDev 7D leads Elo by ~16 days. Games 7D Avg leads by ~21 days. Fight Gap (how long you fight in losses vs wins) shows your mindset shift directly. When you stopped giving up early in losses (Jan 2026+), your rating started climbing.
What Stabilization Looks Like
Overlays compressing toward flat while Elo holds steady or rises = your system is stabilizing. Overlays spiking while Elo drops = destabilization event.
What This Shows
Same Elo trajectory, but the x-axis is cumulative game number instead of calendar date. Each step on the chart is one game — one iteration of the decision-making loop.
Why This Matters
Calendar time is a human unit. Games are the system unit. A week with 50 games and a week with 5 games look identical on a date chart but are wildly different in iterations. This chart strips out calendar drift and shows you the system's true history: decision by decision.
How to Read It
Steeper slopes = bigger Elo changes per game (volatile). Flatter slopes = stable behavior per game. If you compare two visually similar shapes in this chart vs the time chart, you'll see compression more honestly — the same pattern occurring in fewer iterations.
Click any point
Opens the Day Inspector for that game's date. Overlays show behavioral metrics aligned to each game's calendar day.
What This Does
Selects how many games per bucket. Smaller buckets show micro-patterns (session-level). Larger buckets show structural patterns (phase-level).
Trade-off
10 games: ~290 windows, very noisy, see individual sessions
25 games: ~115 windows, session/day patterns
50 games: ~58 windows, multi-day stretches
100 games: ~29 windows, structural phases (RECOMMENDED)
250 games: ~11 windows, macro view
Why Volume Matters
A day with 30 games and a day with 3 games are wildly different at the iteration level, but identical in date analysis. Windowing by games strips out calendar drift — you compare decisions to decisions, not days to days.
What This Shows
The TOP line shows end-of-window Elo at each step. The BOTTOM bars show the Elo Δ for that window (green = window net positive, red = window net negative).
Reading It
Top line steepness: how much Elo moved across that block of N games. A nearly-flat line over 100 games = very stable behavior. A vertical drop = a crash event.
Bottom bars: lets you spot autocorrelation visually. Runs of red bars = compounding losses. Alternating red/green = system finding balance.
What To Look For
The most informative pattern is the bars trending toward zero over time. That means consistent, near-break-even windows — small wins and small losses, no big swings either direction. That\'s structural stability.
What This Shows
Your top 3 windows ranked by Net Elo Δ. These are your highest-performing N-game stretches.
Why Compare Top vs Bottom
The differences between your best windows and your worst windows ARE your performance levers. Whatever varied between them is what you have causal control over. If Best windows have lower volume and higher fight gap and Worst windows have the opposite, that\'s your prescription.
What This Shows
Your bottom 3 windows ranked by Net Elo Δ. Compare these against the Best 3 in the adjacent card to find your behavioral levers.
What This Shows
Every window with all its metrics. Sortable by any column — click a header.
Key Column: Days Spanned
This is the metric you can ONLY see in volume-based analysis. It\'s how many calendar days it took you to play this fixed number of games. Increasing days/window over time = you\'re playing the same N games over longer calendar stretches (less daily volume).
Click a Row
Opens the Day Inspector centered on that window\'s start date.
| # | Date Range | Days | G/Day | Elo Δ | Win% | Drawdown | StdDev | Mv Time | Tempo Gap | Fight Gap | Cycles |
|---|
What This Shows
Your average Elo (left axis, teal) overlaid with the 7-day average of your daily Standard Deviation (right axis, amber). StdDev measures how much your Elo bounced around within a typical day, smoothed over a week.
How to Read It
High StdDev = wild swings within sessions (winning 5, losing 5, up 30, down 40). Low StdDev = tight, controlled sessions where your Elo stays in a narrow band.
Why It Matters
StdDev 7D leads your Elo by approximately 16 days (r = -0.47). When this line compresses downward, your rating tends to rise ~2 weeks later. Your progression: 14.88 (chaos phase) → 7.17 (transition) → 6.48 (recent). This is your system learning session control.
What to Watch For
StdDev compressing while Elo holds = stabilization. StdDev rising while Elo drops = destabilization. If StdDev spikes suddenly, expect rating pressure in 1-2 weeks.
What This Shows
Avg Elo (left) overlaid with the Variance 7D Slope (right). The slope measures whether your variance is increasing or decreasing week-over-week. It's the rate of change of instability.
How to Read It
Green (below zero) = variance is compressing/decreasing. This is good — your system is tightening. Red (above zero) = variance is expanding. Your sessions are becoming more chaotic.
The Key Pattern
When variance slope goes negative (compressing) BEFORE a rating rise, that's your awareness and intention working. This happened clearly in Dec 2025 before your +152 Elo climb. Variance compressed first, then rating followed.
What This Shows
The difference between your average loss game length and average win game length, smoothed over 7 days.
Why It's Here (Honest Framing)
Statistical testing showed this metric is not stable as a predictor — in the first half of your data it correlated positively with next-week Elo, in the second half it flipped negative. So it's demoted from "leading indicator" to "behavioral context."
But you\'ve been intentionally working on this specific behavior (not giving up when down a piece). It\'s the clearest mirror of your mindset shift, even if it doesn\'t reliably move the rating.
How to Read It
Negative gap = giving up in losses faster than playing out wins. Positive gap = fighting longer in losses than wins.
What This Shows
Spearman rank correlation between each behavioral metric (measured today) and your Elo change over the next 7 days. Only metrics with p < 0.1 (statistically meaningful) are shown.
How to Read It
Green bars = positive correlation (when this metric is high, your next-week Elo tends to rise). Red bars = negative correlation (when high, Elo tends to drop). Longer bars = stronger predictors.
Your Top Predictors
Variance 7D Avg (r=-0.34): When your recent variance is high, next week tends to be bad. StdDev 7D Avg (r=-0.32): Same story — instability today predicts losses ahead. Games 7D Avg (r=-0.20): High recent volume predicts lower Elo next week.
| Metric | Calculation | What It Measures | Why It's Here | Predictive Power |
|---|