There are lots of ways to get Kalshi alerts on Telegram: price-move pings, whale-trade trackers, open-source bots. Most are genuinely useful, and almost none of them publish what happened next. We watch every live in-game market on Kalshi at once, fire the second our model and the price disagree on any of them, and then put that call into a graded public log with the losses in it. This page compares the free tools honestly, and then hands you the record.
See the Plans →"Tell me when this market crosses 70¢." Free bots and simple scripts do this well. If a threshold ping is all you need, a free tool is the right answer.
Watch order flow and flag unusually large fills. Good for seeing where money is moving. They tell you that someone traded big — not whether they were right.
Self-hosted scripts that poll the Kalshi API and push to Telegram. Free, flexible, and yours to run — if you're comfortable maintaining code and an API key.
A different job entirely: fire a specific call at a specific price, then log whether it won or lost — publicly. The first three are tools. This one is a record.
| Capability | Free tools (pings, trackers, scripts) | Graded signals (DegenHedge) |
|---|---|---|
| Price-threshold pings | ✓ Excellent — use one | Not the job |
| Volume & whale-trade spikes | ✓ Solid trackers exist | Not the job |
| Flags when the price disagrees with a live model of the game | ✗ No game-state model | ✓ Model % vs. Kalshi price, re-checked every 30–60 seconds |
| Grades its own calls | ✗ Nothing to grade — a ping isn't a call | ✓ Every alert logged win or loss |
| Publishes losses | ✗ | ✓ All of them, in the results log + CSV |
| Pre-commitment proof | ✗ | ✓ Locked Telegram posts before settlement |
To be clear about the first two rows: if all you want is "ping me when BTC-hourly crosses 70¢," a free bot does that perfectly and you should not pay anyone for it. What a threshold ping cannot tell you is when a live model of the game disagrees with the price. That takes a model of the game state running next to the market. We are careful about what that is worth: we do not claim the price is wrong, and we do not claim it will move toward us — our own order-book instrumentation found no measurable move after an alert, which we write up at /kalshi-lag/. What we claim is the detection, delivered fast, and published to a log that is graded at settlement with the losses in it.
The moment a signal fires, the timestamp is recorded. No back-dating, no "we called it earlier."
The model's win probability and the live Kalshi price of the called side are logged at fire time — the two numbers the whole call rests on.
When the market settles, the result goes in the log. Losses are never deleted, reworded, or quietly dropped.
Sortable at degenhedge.com/results/, every loss included, plus a full CSV download at /results.csv.
MLB and NFL calls post to t.me/DegenHedgeProof: MLB as a locked post ~10 minutes after each fire (side hidden), NFL 12 minutes after each fire with the side shown, both graded at settlement, losses included. Running since June 13, 2026. BTC calls post there as graded recaps every 3 hours after every window in the 3-hour block has closed, never live.
The current MLB record under V2.2 filters: 948 first-signal games graded — 748 wins, 200 losses (78.9%, at a median entry of 80¢), from 1,381 logged signals. Four entries straight from that log:
MIN @ CLE — bottom 11th, MIN down 1. Model: 100% on MIN · Kalshi: 75¢. Final: MIN 2–1.
NYM @ WSH — bottom 9th. Model: 90% on NYM · Kalshi: 65¢. Final: NYM 2–1.
DET @ CWS — bottom 10th. Model read 100% on DET at 79¢. CWS walked it off 4–3. Even certainty-grade model reads lose.
HOU @ KC — 4th inning. Model: 89% on KC · Kalshi: 66¢. HOU came back to win 8–7.
Grading everything teaches you things a highlight reel never would. Example: a wider model-vs-price gap is not a safer signal. When we bucket the graded set by how far the model sat from the price, the widest bucket does not win more often than the narrowest, and the middle bucket wins least of the three. That is why the gate is capped at the top as well as the bottom, why the difference is not a field in an alert, and why we do not publish it as a number: when the market disagrees with the model by more, part of that disagreement is information — injuries, bullpen state, weather — that the model doesn't see. The record that matters is the whole graded set, 78.9% at a median entry of 80¢, and every row of it is in the public log. What "graded" means in full depth — grading rules, filters, edge cases — is at /kalshi-signals/.
/teams/teams/manage in the bot, or email [email protected]. Cancelling later stops future renewals at the end of the current paid month and does not refund it. MLB only: MLB alerts run through the World Series in late October. The feed is quiet over the winter. After the final World Series game we pause billing on MLB subscriptions bought on degenhedge.com, so nothing is charged over the winter. Billing resumes shortly before Opening Day, and we email you at least 7 days before the first charge. Cancel any time from your account page. No refunds except as required by law. By starting a trial you authorize these recurring charges and agree to the Terms and Privacy Policy. Signals are informational and for entertainment only, are not financial or betting advice, and any single signal can lose 100% of your stake. Must be 18+ (some platforms require 21+). If gambling is a problem for you, call 1-800-GAMBLER.
Every signal can lose. About 1 in 5 graded MLB games is a loss, and losses cluster — the log shows losing streaks, not politely spaced-out misses. A model read of 100% has lost (see DET @ CWS, May 30, above). If a run of 5 straight losses at your chosen size would hurt, the size is wrong.
This is entertainment, not investment. The model is a fast state reader, not an independent predictor of games, and against settled outcomes it runs optimistic. Past performance does not guarantee future results. You must be 18 or older. Only risk money you can afford to lose.
Read the full model details →The model re-checks every tracked game and market roughly every 30 seconds; when a signal fires, the Telegram message goes out within seconds. Speed is what you are paying for: a live game state changes constantly, so a detection is only worth anything while it still describes the game in front of you, and BTC windows only live for 15 minutes. Speed is not a claim that you are beating anyone to a price move — we tested that and it did not hold.
A free Telegram account. Subscribe via Stripe, connect the bot from your welcome email, and alerts arrive as ordinary Telegram messages. A Kalshi account is only relevant if you decide to act on a signal — that decision is yours, not ours.
Anytime, through the Stripe customer portal (linked in every receipt) or by emailing [email protected]. Access runs through the end of the current billing period.
Don't take this page's word for it. The full sortable log — every loss included — is at /results/, the raw data is at /results.csv, and the pre-commitment channel at t.me/DegenHedgeProof shows MLB and NFL calls posted before settlement, graded win or lose, plus BTC graded recaps every 3 hours.