How to Find Your Own Betfair Tips

Q: What's the easiest sport to start with?

UK and Irish horse racing has the most freely available form data, deep liquidity, daily volume, and short payoff cycles — ideal for learning a research method.

This article is a sub-article in the Betfair Daily Tips and Trading Ideas pillar. The pillar covers the daily tip ecosystem; this one drills into the research method behind finding your own tips. If tipster economics is what brought you here, start with are tipster services worth paying for?

Why "Doing Your Own Research" Beats Paying for Tips

Three structural reasons:

Cost is zero. The data is free. You pay only your time.
You learn while you work. The 100 hours you'd spend reading tipster picks teaches you nothing. The 100 hours you spend pricing matches yourself builds an actual edge.
Your edges are uncorrelated with the crowd's. Public tipsters move prices. Your picks don't.

The Five-Step Workflow

Step 1: Choose your domain

Pick one sport and one or two competitions you watch anyway. A workflow across 6 sports produces 5x the noise and 1/6 of the depth. Default options for retail traders:

UK & Irish horse racing — daily volume, deep liquidity, great free form data.
Premier League and Championship football — deep public data, mature markets.
ATP and WTA tour tennis — clean data, liquid in-play.
IPL and international cricket — high IPL volume, season-bounded.

Avoid: niche sports until you can audit yourself; multi-sport scanning workflows.

Step 2: Build a fixed daily data pull

Same sources, same time, same order. Habit beats ambition. For Premier League:

FBref — xG, xA, form last 8 matches.
Football-Data.co.uk — raw results, home/away splits.
Premier League official — injury news, suspensions.
Betfair Hub Football — current price screen + analyst note.
Twitter/X xG community — late-week analytics threads.

For UK racing: Timeform free pages, Racing Post free morning preview, ATR/Sky form, Betfair Hub racing card.

Step 3: Price each match yourself BEFORE looking at Exchange

This is the discipline that separates research from rationalisation. Write down a probability for each main outcome before you see what the Exchange is offering. If you see the Exchange first, your "estimate" will silently anchor on it and you'll find phantom value.

Use whatever model fits your sport: Elo, xG-based regression, surface-adjusted serve-return, par-score for cricket. The model only has to beat the market on the margin.

Step 4: Compare to Exchange prices and compute edge

Pull current Exchange prices on the markets you priced.
Compute commission-adjusted edge. Method: value bets: finding edges today.
Flag only selections with edge ≥ 3-5%.
Note potential in-play follow-up plans (where would you green up? where would you stop?).

Step 5: Execute, log, review

Place positions at the size your bankroll allows. Log entry price, stake, fair price, edge, in-play decisions, exit price. Weekly review: is realised return tracking your computed edge? If not, your model is wrong somewhere — find where and fix it.

Worked Example — Saturday Premier League Workflow

Saturday morning, Premier League afternoon kickoffs

8:00 — Data pull (30 min): FBref last-6 xG-for / xG-against for both teams in each fixture. Premier League injury news. Football-Data home/away splits.

8:30 — Probability estimate (20 min): For each fixture, project combined xG. Use 3-year mean Match Odds correlation table to convert combined xG to win/draw/loss probabilities and Over/Under and BTTS rates.

8:50 — Fair price table (10 min): Convert probabilities to fair prices on Match Odds, Over/Under 2.5, BTTS. Round to tradeable ticks.

9:00 — Cross-reference Exchange (10 min): Open Betfair. Note current prices. Compute edge on each market. Flag selections ≥4% edge after commission.

9:10 — Plan execution: Note pre-match position sizes and in-play setup triggers (e.g., "Lay Draw at 3.20 on Brighton v Tottenham; hedge at 5.5 on first goal").

14:30 — Execute pre-match positions, set in-play triggers in Bet Angel, watch matches.

23:00 — Log results.

The Data Sources You Need Per Sport

Horse racing

Racing Post free pages and morning preview.
Timeform free profile pages.
Sporting Life racecards.
Attheraces / RUK pre-race shows.
Betfair Hub racing card.

Football

FBref — xG and advanced stats.
Understat — quick xG comparison.
Football-Data.co.uk — raw results, free download.
Premier League / Bundesliga / La Liga official pages — injury and lineup news.
Twitter/X xG analyst community.

Tennis

Tennis Abstract — serve %, return %, surface splits.
ATP / WTA official sites — schedule, draw, head-to-head.
Match Quality and Elo feeds on Twitter/X.

Cricket

ESPNcricinfo — previews, lineups, ball-by-ball.
Cricsheet — open ball-by-ball historical data.
CricViz — predictive numbers (some paid samples).

What to Track in Your Bet Log

Date, sport, event, market, selection.
Pre-match: your probability estimate, your fair price, Exchange price taken, computed edge.
Position: stake, in-play exit plan, stop level.
Result: matched in/out, P&L, commission paid.
Notes: what surprised you, what you'd change.

Without a log, you can't tell signal from variance. With a log, after 100 bets you can see where you have edge and where you don't.

Common Beginner Mistakes

Skipping the price-before-Exchange step. The Exchange price will quietly anchor your "estimate". Discipline: model first, look at the market second.
Researching every sport, mastering none. Depth beats breadth.
Trusting tipsters as "second opinions". Their opinion contaminates your model. Read tipsters AFTER you've made your pick.
Logging only the wins. Selection bias destroys evaluation. Log all.
Building a "model" with no historical backtest. If you can't show 200 historical matches that your model would have priced correctly, it's not a model.
Letting variance change the model. Bad result ≠ bad pick. Audit the process, not just outcomes.

Software That Helps

Our free trading calculator — fair price, dutching, hedging, green-up.
Bet Angel — spreadsheet automation; you can wire your fair-price model into automated entry triggers.
Geeks Toy — fastest execution once you've flagged opportunities.
Cymatic Trader — free, sufficient for beginners.
Betfair API — for advanced traders building automated models.

Building Daily Habit Around Research

The biggest predictor of success in self-research is not analytical talent — it's daily habit. Two principles that double sustainability:

Same time, same place, same tools. Set up a desktop with browser tabs pre-arranged: FBref, Football-Data, official league sites, Betfair, your spreadsheet. Friction kills compliance.
30-minute minimum sessions. 30 mins is enough to score 4-5 fixtures properly. Less than 30 leaks rigour; more than 60 invites overthinking.

The Three Sport Workflows in Detail

Football workflow (Saturday Premier League)

Friday evening: pull FBref last-8 xG-for/against for each team in Sat fixtures. Note injury press from official sites. Saturday 08:00: lineups not yet posted — write provisional probabilities. 12:00: refresh injury news, adjust. 14:00 (60 min pre-kick): confirmed lineups. Adjust for surprise inclusions. 14:30: open Betfair, compute edge, place pre-match positions, set in-play triggers.

Tennis workflow (ATP/WTA daily)

Evening before: open day's draw on ATP/WTA. Pull Tennis Abstract surface splits for each match. Check schedule for back-to-back fatigue. Note conditions for each court (outdoor/indoor, surface speed). Morning of: write fair prices. 2 hours pre-first-serve: cross-check Exchange. Place positions pre-match.

Horse racing workflow (UK daily)

Morning preview: open Racing Post free preview pages. Note any non-runners. Build your tissue prices per race. 30 min before race: check current Exchange. Compare to your tissue. Identify drift candidates (long horses), steam candidates (shortening horses). Trade at the off, exit in-running based on pre-race plan.

Worked Example — Friday Research, Saturday Action

Friday evening setup, Premier League Saturday

Fixture A — Top half v top half, both fit. Combined xG 3.2. Probabilities: Home 42%, Draw 26%, Away 32%. Fair Match Odds: 2.38 / 3.85 / 3.13.

Fixture B — Mid v relegation, away missing key CB. Combined xG 2.8 but home form weak. Probabilities: Home 38%, Draw 28%, Away 34%. Fair Lay Draw: 3.57.

Saturday 14:00 Exchange check: Fixture A — no clear edge across markets, skip. Fixture B — Lay Draw at 3.40, your fair is 3.57. Edge +5.0% post-commission.

Action: Lay £100 of Draw in Fixture B at 3.40 (liability £240). Plan: hedge at 5.0 on first goal for ~£32 green; stop at 2.80 if 0-0 with no shots at HT.

The Probabilistic Mindset

Your job as a researcher is to assign probabilities, not predict outcomes. A 60% favourite loses 40% of the time. You will be "wrong" plenty. The question is whether your probabilities, over thousands of bets, are calibrated:

Of selections you priced at 60% probability, did roughly 60% of them win?
Of selections you priced at 30%, did roughly 30% win?

If yes, you have a calibrated model. If your "60%" selections won 45%, your model is overconfident; if 75%, underconfident. Either way, recalibrate.

How to Avoid the "Big Spot" Trap

Big spots — Champions League finals, Grand Slam finals, World Cup matches — are the worst places to deploy your research. They're the most efficient markets on the Exchange because every professional, every model and every media outlet is focused on them. Your free FBref xG numbers add nothing the market doesn't already see.

Spend big-spot weekends on mid-table league fixtures or lower-tier tour matches. That's where the public attention deficit creates retail-trader edge.

The First-Month Research Plan

For your first 30 days of doing your own research, aim for:

Days 1-7: Pick your sport. Set up your spreadsheet. Practice pricing 10 fixtures without placing bets. Compare your prices to Exchange closing prices.
Days 8-14: Continue pricing without betting. Track which markets your prices best beat the closing line on.
Days 15-21: Place small live bets (£5-10) on your highest-edge picks. Log every detail.
Days 22-30: Increase to standard stake size. Review week-over-week.

Resist the urge to skip to live betting. The first weeks build calibration; missing them is why most amateur researchers fail.

Tools to Speed Up Research Without Cheating

Notion or Obsidian for tracking research notes, hypothesis tests and bet rationales.
Google Sheets with IMPORTHTML to pull free data programmatically.
RSS readers for injury news from official sources.
Discord communities like the trading Discord servers for sanity checks (after you've made your own picks).
Browser bookmarks bar with your six data sources, in your daily-pull order.

Frequently Asked Questions

How long does daily research take?

For a focused single-sport workflow, 45-75 minutes per day, plus 15 minutes of execution. For multi-sport, expect 2-3 hours.

What's the easiest sport to start with?

UK and Irish horse racing. Daily volume, deep liquidity, free form data, short payoff cycles, fast feedback loop.

Do I need software to find my own tips?

No, but a calculator and a logging spreadsheet are mandatory. Trading software (Bet Angel, Geeks Toy) becomes valuable once you're executing 5+ positions a day.

How do I avoid bias when pricing my own picks?

Write probabilities down BEFORE looking at the Exchange. Use the same model the same way each day. Backtest historically. If your in-the-moment numbers differ wildly from your historical model output, you've drifted — recalibrate.

Should I tell anyone my picks before posting?

No. Public posting accelerates price movement against you. Keep picks private until you've executed.

Honest Risk Note

Doing your own research doesn't guarantee profit — it just removes the structural cost of subscriptions. Most retail bettors lose because they don't have an edge, not because they paid for tips. Audit yourself honestly. BeGambleAware.org if betting is causing distress.

How to Find Your Own Betfair Tips — A Research Method