Transparency World Cup NFL vs. Touts 🎓 Academy Join Free on Telegram →
Module 1 of 8 · ~2 min
What is DOGINME HQ?
Let's clear up what this platform is — and isn't — in under two minutes.

❌ It is NOT:

❌ Not this
A sportsbook or betting platform. We don't take bets.
❌ Not this
A picks group. No human is posting "locks" to a channel.
❌ Not this
Gambling advice. Results are data — what you do with them is your choice.
❌ Not this
A black box. Every signal, every result, every stat is auditable.

✅ It IS:

✅ This
Sports Intelligence Platform. Machine-generated signals across 8+ sports.
✅ This
Verified Results. WIN/LOSS auto-resolved from official box scores — not self-reported.
✅ This
Transparent Performance. All-time record, tier breakdown, CLV — public and auditable.
✅ This
Tracking Tool. Log your picks, track your edge, see your personal win rate over time.
The one-sentence version
DOGINME HQ is a machine-powered sports signal engine with auto-verified results, a transparent performance record, and a community of people tracking their edge — all inside Telegram. Free.
Module 2 of 8 · ~2 min
Understanding Signals
Every signal the platform generates looks like this. Tap each field to learn what it means.
⚾ MLB
Tier: SNIPER
Player: James Wood — 1+ Hits
Line: -150   Edge: +14.2pp
Signal ID: #ab12cd34
Generated: 2026-06-10 09:14 ET
Tier — Signal Confidence
The model groups signals into tiers based on edge size and confidence. SNIPER is the highest-confidence tier — reserved for signals where the model has the strongest statistical edge. There's also STRONG (high confidence, high volume) and CORE (medium confidence). Fewer SNIPER signals, but historically the highest win rate.
Player Prop
This signal is on James Wood getting 1 or more hits in today's game. The model analyzed his historical hit rate, today's pitcher matchup, batting order position, park factors, and weather. The result: the model believes the market is mispricing this prop.
Line — Market Odds
-150 means you'd need to risk $150 to win $100. This is the implied probability the market assigns (~60% chance of 1+ hits). The model's own probability estimate is higher — which creates the edge. You can compare this against any sportsbook showing this prop.
Edge — Model vs Market Gap
+14.2pp means the model estimates the true probability is 14.2 percentage points higher than what the market is pricing. If the market says 60% and the model says 74.2%, that gap is the edge. Bigger edge = stronger signal. Edges below 3–5% are typically noise.
Signal ID — The Immutable Record
Every signal gets a unique ID the moment it's generated. You can use this ID to run /wincard ab12cd34 in the bot at any time — even months later — to see the full signal record: what was predicted, what happened, the final CLV, and the auto-resolved outcome. It cannot be edited.
💡 Try it: tap each highlighted field above to see what it means. Every field in a real signal has this explanation available in the bot via /help signals.
Module 3 of 8 · ~2 min
The Tracking Loop
This is what separates DOGINME HQ from a normal picks channel. The full lifecycle of a signal — start to verified result.
📡
Signal Generated
The scanner finds a mispriced market. Edge is calculated. Signal is published instantly to the bot.
✅
/took <signal_id>
You tell the bot you're playing it. Your pick is logged — timestamped and immutable. No editing later.
🔄
Resolver Runs
After the game, the resolver pulls official stats from the BallDontLie API or official match data. Outcome is calculated automatically — no human touches it.
🏆
Win Card Delivered
You get auto-DM'd a Win Card (or Loss Card). It shows the signal, the result, the raw stat used, and the data source. Verifiable against the official box score.
📈
CLV Calculated
The model compares your entry price to where the market closed. This is your Closing Line Value — a measure of whether you beat the market, not just the outcome.
/profile Updated
Your personal stats update: W/L record, win rate, streak, CLV average, rank score. Run /profile at any time to see where you stand.
Module 4 of 8 · ~3 min
The CLV Academy
Closing Line Value is the most important concept for serious bettors. Let's learn it in 3 minutes with no math jargon.
📊 The Scenario
You bet James Wood 1+ Hits. At the time you placed it, the odds were:
Your entry price
-150
Market close price
-185
The market later moved to -185 — more expensive — before the game started.
Question: Did you beat the market?
✅ Yes — you beat the market.
You locked in -150 before the market agreed the bet was worth -185. You got a better price than what the sharpest money eventually settled on.

Even if James Wood didn't get a hit, you still had positive CLV. You made the right decision. Short-term results are noise — long-term CLV is signal.
Why does CLV matter?
Win rate fluctuates. Variance is real. But consistently getting better prices than the market close is the only mathematically sustainable edge. Professional bettors track CLV obsessively — it tells them if their process is working, independent of short-term luck.
What does DOGINME HQ track?
Every signal's CLV is calculated after market close. Your personal average CLV is shown in /clv. The model wall shows average CLV per sport and tier. STRONG tier averages +8.4pp CLV — meaning on average, the signals beat the market close by 8.4 percentage points.
CLV = Closing Line Value
+CLV = Beat the market
-CLV = Behind the market
/clv for your personal stats
Module 5 of 8 · ~2 min
The Transparency Layer
We don't ask for trust. We show the data. Live numbers directly from the platform archive — updated every 2 minutes.
Loading live data...
...
Total signals generated across all sports
...
Resolved signals with verified outcomes
...
Platform-wide win rate (all sports, all tiers)
...
Average CLV on resolved signals (STRONG tier)
How to verify it yourself
/wincard <id>
Full immutable record for any signal. Raw stat, data source, resolver output.
/accuracy 30d
Platform accuracy wall for last 7, 30, or 90 days. Tier breakdown included.
/simulate
Run a bankroll simulation over the full archive. Every signal, wins and losses.
/modelwall
Sport × tier performance table. N, wins, win rate, avg CLV. Live numbers.
Module 6 of 8 · ~2 min
World Cup War Room
FIFA World Cup 2026 runs June 11 – July 19. DOGINME HQ is covering every match. Here's how to use the WC war room.
Loading WC data...
Loading live WC signals...
WC commands
/wc
Live WC signals for today's matches. Tier, edge, kickoff time.
/wc history
All resolved WC signals. Auto-verified results from official match data.
/wc rank
WC leaderboard. See where the pack stands on tournament picks.
/took <id>
Log a WC pick. Get auto-DM'd when the match resolves.
Module 7 of 8 · ~2 min
DOGINME HQ Labs
This is our moat. Real model improvements, real audits, real lessons — documented publicly. No competitor does this.
Case Study 01 — June 2026

⚾ The K-Model Direction Bias

The strikeout (K) model was selecting UNDER 92% of the time — a systematic bias. Outcome resolution was always checking the OVER direction first, masking the real win rate. We identified the bug, added selected_side tracking, and fixed the resolver.

Result: Bias dropped from 1.394 → 0.909. Combined WR improved +1.2pp after fix.
Case Study 02 — Ongoing

📊 MLB Hits Model Calibration

The SNIPER tier (highest confidence) has an 8× higher win rate than STRONG on HR props — but only 8 signals. We track day game performance separately because day games show a −15.4pp win rate gap vs. night games. The model adjusts for game time, altitude, park factor, and pitcher fatigue simultaneously.

Current: MLB SNIPER 62.5% WR (n=8) · STRONG 59.6% WR (n=1,259)
Case Study 03 — Storage Architecture

🔧 Persistence-Backed Signal Archive

Early versions kept resolved signals in memory. On restart, the entire record was lost. We rebuilt the storage layer: every signal, ack, and resolution is now persisted to disk and reconciled on startup. Signal IDs are immutable. The archive goes back to the first signal generated.

Result: 3,388 signals in archive. Zero data loss since persistence rewrite.
🔬 Why we publish this: Any platform can claim a win rate. Few publish the bugs they found and fixed. Labs builds long-term trust — if the model is wrong, we say so publicly and show how we fixed it.
Module 8 of 8
Graduation
You've completed the DOGINME HQ Academy. Here's what you know that most users don't.
🎓
DOGINME HQ Academy Graduate
Completed June 2026
✅ What DOGINME HQ is (and isn't)
✅ How to read a signal: tier, line, edge, ID
✅ The full tracking loop: signal → /took → win card
✅ What CLV is and why it beats win rate alone
✅ How to verify any result yourself
✅ How to use the WC war room
✅ Why Labs transparency is our moat
🚀 Launch DOGINME HQ on Telegram →

Free. No card required. Works on iPhone, Android, and desktop Telegram.

Your first 3 commands in the bot:

1. Run /wc — see today's World Cup signals
2. Run /took <signal_id> on any pick you're playing
3. Run /profile after your first resolved pick
1 / 8