
1,024 installs got it into the top 30 of Turkey's sports Top Grossing chart: Badiscore didn't blow up, it picked a chart with a low bar
What it is A pre-match probability prediction and live score app for soccer and basketball. Web, iOS, and Android share one account. Each match gets three predictions under public rules, and all results stay on the track-record page after the game. Pro is ₺499.99 a month.
Why it took off Turkish users were already willing to pay for "match predictions," and the local sports chart has a low bar. It wrote its store title in "Maç Tahminleri," the most common local search term, and a few subscriptions plus about a thousand installs put it on the chart.
CHow you could follow Don't head-to-head in Turkey. Find a smaller-language market where people also pay for predictions and the sports chart has a lower bar, and build on the same structure (odds converted to probabilities, rules for choosing predictions, a few free matches a day, subscriptions sold on the web). From day one, keep a public track record that never deletes losses, and use high-frequency local keywords in the title.
Est. monthly revenue (Google Play) under $1KWe estimated this from its rank on country top-grossing charts and real install counts (it isn't on the grossing chart, so confidence is lower), calibrated against real revenue; typical error is 2–3x. Read it as a tier, not an exact number.
- Data (updated daily) · first launch 2026-10-27 · 11-16 installs 1,165
Android · Total installs worldwide Data 2026-11-01 → 2026-11-16 (16 days, 15 observations). Google's detail page only refreshes install counts every other day, so a flat day doesn't mean nobody installed.
Badiscore is a soccer and basketball pre-match probability app built solo by an indie developer in Ankara. It launched on iOS in Oct 10 and on Google Play in Oct 27. In November it sat at #16 to #30 on the iOS Turkey sports Top Grossing chart, but Play has only 1,024 installs worldwide, and the iOS Turkey storefront has 7 ratings in total. It charted because Turkey's sports category has a low bar, the title uses the words locals search for most, and Turkish bettors were already willing to pay for "match predictions." What sets it apart from others is that it never deletes losing predictions. That earned it good reviews, but it didn't bring in new users. Below: how it works, where the money comes from, how crowded this niche is, and whether you should follow it.
"It keeps the losses too"
The European Nations League match on Nov 12 evening, Turkey at home, lost 1–2 to Belgium. Before kickoff, Badiscore's Performance page had three predictions for the match, each with a probability: the lead pick "Belgium to score over 2 goals" at 58%, the second pick "Turkey win neither half" at 54%, and the bold pick "Belgium win and total goals over 2" at 42%. After the match, the lead pick and the bold pick both hit.
A few rows up on the same page is Montenegro 2–1 Latvia on the same day. Its lead pick was "both teams not to score" at 61%. Both teams scored, so that row shows a dash, meaning a miss. Under the page title there's one line of small text: misses stay on the record too.
That's what Badiscore wants to sell you. On the iOS Turkey storefront in Nov 1, a five-star reviewer named Santazmo wrote: "Diğer tahmin uygulamaları gibi geçmiş tahminleri de silmiyorlar", meaning it doesn't delete past predictions the way other prediction apps do. In Turkey's "tip" communities, deleting losing picks and posting only the wins is a longstanding habit.
One person, one month, a website and two apps
The developer is Hasan Metehan Bacı, listed on Play as metekog, based in Ankara. According to his personal site and LinkedIn, he has seven-plus years of experience in fintech and defense simulation. The timeline is tight:
| Date | Event |
|---|---|
| 2025-12-29 | Opened the X account @badiscore |
| 2026-09-29 | Registered badiscore.com |
| 2026-10-10 | iOS release |
| 2026-10-27 | Listed on Google Play |
| 2026-11-11 | iOS updated to 1.8.11 |
From Oct 19 to Nov 11, the iOS version shipped roughly 10 releases, and Play was updated almost daily. That one person can iterate this fast suggests heavy use of AI coding (inference).
The product is "pre-match probabilities + live scores." Soccer covers 35 countries' 48 leagues; basketball has the NBA, EuroLeague, and EuroCup, plus domestic leagues in six countries: Spain, Italy, Germany, France, Greece, and Turkey. Soccer gets three predictions per match, basketball one. Web, iOS, and Android share one account, Pro membership works across all three, there are no ads, and it's for users 18 and up.
It also explains how the probabilities are produced. Outcome and win/lose markets are converted directly from pre-match odds; corners, yellow cards, shots, and saves use its own model. How the three predictions are chosen follows a public rule set: first decide "how the match will play out" (if the favorite's win rate is at least 45%, expect the favorite to win, otherwise look at total goals), then pick options that hold together at 85% or higher when that expectation holds. Options with probability above about two-thirds are skipped, as are odd/even and half-time markets. The probabilities are recalculated only once, after the starting lineups are announced, and only if they shift by more than 8 points. That way the three predictions tell the same story, and if the read on the match is right, all three can hit together.
How low the bar is for chart placement
Start with Google Play (cumulative global installs):
| Date | Total installs | Turkey sports free chart |
|---|---|---|
| 11-01 | 1 | 121 |
| 11-03 | 171 | 128 |
| 11-08 | 361 | 122 |
| 11-11 | 460 | 46 |
| 11-12 | 460 | 38 |
| 11-13 | 1,024 | 46 |
The install count climbs in steps (+170, +167, +564) with flat stretches in between. Google likely updates this number in batches, so don't read too much into single-day changes. It only charted in the Turkey store, but the listing page also opens in the US store, so it isn't limited to Turkey.
iOS Turkey sports Top Grossing: 10-28, #88,11-03, #51,11-10 dropped to #135,11-11, jumped to #16,11-13, #30; its best Top Free rank was #11-11 in 21. In Azerbaijan, a Turkic-speaking country, it reached #8 on the Top Grossing chart at one point (10-30). The iOS Turkey rating count went from 0 at the end of October to 11-08 with 7 ratings, and hasn't moved since. All are 5 stars.
These ranks don't tell you much. Comparing on the same charts: on Play's Turkey sports Top Free, #46 takes only about 1,000 installs; on iOS Turkey sports Top Grossing, #22 Rega AI has only 11 ratings and #23 Bowken only 2. On Turkey's sports Top Grossing chart, a few subscriptions are enough to push a new app into the top 30.
Where users come from
- Store search (confirmed). On the Play Turkey store the title is "Badiscore: Maç Tahminleri", which means "match predictions". That is the term locals actually type when they search for a prediction app. The descriptions are all in Turkish, opening with the Süper Lig and the Turkish Cup. The iOS subtitles also differ by store: the Netherlands store uses Turkish "Canlı Skor & Tahmin" (live scores & predictions; presumably aimed at Turks living in the Netherlands), the Swiss store uses German "Fußball-Prognosen", and the US store uses "Soccer Predictions". Almost all the charting happens in the Turkey and Azerbaijan stores.
- Low-bar category charts (confirmed). As noted above, charting gets you a bit more exposure, and more exposure brings a few more sales.
- Website SEO (in place, effect unmeasurable). badiscore.com's sitemap has 20,784 URLs: 8 languages multiplied by page types for teams, matchups, leagues, and basketball. These are programmatically generated pages. Every page carries a "Get the app" QR code, and the download links are tagged with their source. Chrome real-user data shows that October, a domain registered just one month earlier, already ranks in the top 500k of Turkish websites.
- Possibly ran ads (unverified). The release notes for versions 11-05 and 11-07 added an "ad measurement" toggle, which suggests an ad attribution tool was integrated. Right after that, from 11-10 to 11-11, the iOS Turkey, iOS Azerbaijan, and Play Turkey charts all jumped at once. Those days fell during an international break with few matches, so it doesn't look like organic traffic from matches. Still, I couldn't find any of its ad creatives, so from the outside there's no way to tell whether it ran ads or how much it spent.
- Own social media (present, but tiny). 8 posts on X; on TikTok @badiscorecom has 7 followers and 21 videos, each getting 136 to 550 views; on Instagram it does NBA breakdowns.
- Press and communities: none. Search turns up no coverage, no Reddit threads, and no forum discussion, and nobody on X mentions it.
Where the money comes from: ₺499.99 a month, and Turks are already willing to pay for this
App Store Turkey in-app prices (checked 11-13): Badiscore Pro is ₺499.99 monthly and ₺2,999.99 yearly, which at current exchange rates is roughly $10+ a month and about $70 a year (estimate). New users can try it free for 7 days. On Play the listed IAP range is $9.99 to $59.99. You can see scores and the standings without paying. Registering a free account unlocks the full content of 2 football matches and 2 basketball games a day (1 of them NBA); everything else requires Pro. Its website also takes subscriptions directly, and the site carries the pre-sale disclosure and distance-sales contract that Turkey's e-commerce law requires. Revenue collected on the web doesn't pay the store's commission.
It can charge this price from day one because the demand was already there. The top free sports apps on the Turkish Play store are licensed football betting-coupon (iddaa) platforms such as Nesine, iddaa.com, Bilyoner, and Misli. Among the top 80 iOS Turkey sports grossing apps, roughly 30 are prediction apps. In Turkey, paying for "predictions" and "tips" is an established category. Badiscore hit the iOS Turkey sports grossing chart on day 18 of launch (10-28), which suggests some of its first users were already paying.
How much it actually earns is not public and can't be calculated. Given that it reached the top 30 of the grossing chart with only 7 ratings, monthly revenue is most likely still small (inference).
Who it's competing with
This niche is extremely crowded in Turkey. In 2026 years, at least ten new AI prediction apps from Turkish indie developers have launched.
| Product | Launched | Scale | How Badiscore is different |
|---|---|---|---|
| Akıllı Saha | 2026-09-26 | No. 4,51 sports grossing in iOS Turkey by rating count; 10,837 Play installs | Same period, same market, an order of magnitude bigger |
| TipsPro AI / BestTips | 2024-09 | iOS Turkey grossing No. 6,386 by rating count; Play 326k | Sells recommended slips, does not publish odds or track record |
| Scouter | 2023 | Play 1.12M; iOS Turkey grossing No. 13 | Multi-country prediction tipster app |
| GoScore AI | 2021-12 | Play 1.1M | Long-running local Turkish player |
| 11XScore AI Football | 2025-08 | Play 3.02M, +7k in the last 54 days, charting in 25 countries | Heavy paid acquisition |
| 180Score | 2024-03 | Play 2.93M, IAP $3.99–59.99 | Nearly identical pricing, rolled out globally |
| Forebet / BetMines | 2014 / 2019 | Play 3.57M / 7.15M | Long-established odds site with ads |
| FootyStats | Around 2017 | iOS Turkey grossing No. 21 | Sells statistical data, with Turkish-language prediction pages |
| Goal Signal, QuantGoal, Rega AI, Kahin AI, Tahmin Makinesi, and others | 2025-12 to 2026-11 | Ratings in the single digits to dozens; Kahin AI has 1,362 Play installs | Same wave as Badiscore, roughly the same size |
| Mackolik, Sofascore, Flashscore, plus editorial picks from Nesine and Bilyoner | Since 2010 | Tens of millions to over 100 million installs | Odds, polls, and editorial picks are all free |
Chart positions, ratings, and install counts are as of around Nov 13.
What is actually different about it, and why it hasn't scaled
There are three real differences. First, the track record is checkable: every prediction stays on the Performance page after the match, losses included. It also publishes a calibration set: picks given a 60% to 65% probability hit 62.5% of the time across 703,127 options, and 59,414 basketball analyses hit 61.5%, against an expected 60.9%. It says that before launch the rules were backtested on historical matches, so a good share of those numbers are probably backtests rather than live results (inference). Second, three predictions come from one judgment and don't contradict each other. Third, one account across three platforms, no ads, covering both football and basketball.
That explains why it earned good reviews, but not where its new users come from: 4 reviews all 5 stars, praising "analizler çok iyi" (the analysis is great), the win/loss streak page, and the fact that it doesn't delete past predictions. New users come from local keywords and low-barrier charts. The foundation is pre-match odds, plus a not-too-complex in-house model that anyone in the field could build in a week or two. Its only moat is the public track record it has already built up, and that takes time to accumulate.
If it were you
C: worth learning from, but following it isn't a good use of time. Badiscore itself hasn't broken out yet: 1,024 installs and 7 ratings, with its rank coming from low-bar charts. The Turkish prediction market it sits in already has about thirty comparable products, and a dozen new indie developers have shown up in a year. There's no technical barrier; anyone can build the same thing. What's worth taking is three tactics, not this product or this market.
Who it suits: people who understand a small-language sports market, either as native speakers or because they live there, who know which words locals use when searching for predictions and buying tips, and who know which claims are prohibited there. They should be able to build a data pipeline, convert odds into probabilities, and be willing to publish automatically every day and make the track record public, with a team of 1 to 2 people.
Who it doesn't suit: anyone hoping to acquire users on the words "AI predictions" alone. In a market like Turkey, the front of the pack is made up of 300k to 3M-install products that buy globally, plus licensed betting platforms that give away tips for free. It's also not for anyone aiming at English or Spanish-speaking markets. And this category borders on gambling: app store review, ad placement, and payments are restricted in many countries, so stay away if you don't want to deal with that.
How to do it: don't fight head-on in Turkey. Pick a small-language market that also pays for predictions and has low-bar sports category charts. First count how many prediction apps are on the local store's sports grossing chart and how many ratings the No. 20 app has, then decide whether to go ahead. You're looking for a market where a few dozen ratings are enough to reach the top 20. Build the product on the same structure: convert pre-match odds into probabilities, pick predictions by fixed rules, give away a few matches free each day, and sell monthly and yearly plans. Use one account across web and app, and collect subscriptions on the web. From day one, publish every prediction alongside its result and never delete any. The earlier you start, the sooner the track record builds up, and that record is your moat.
How to get users: put the term locals search for most in the store title, write the description in the local language, and open with the local league. On the website, generate pages in bulk by "team, matchup, league × language" and put a download QR code on each one. On short video and X, rather than posting "guaranteed win today", post a review of the previous day's three predictions, right and wrong, so that the fact that misses are kept on record becomes visible.
When to move: don't rush. The window isn't tied to a date. It depends on whether someone in the market you chose has already started building a public track record. If you do want a target date, aim for the start of a major season (European leagues kick off in August, or after the January transfer window in 2027), so the track record page starts recording from the first round of the season.
What would show this read is wrong: ① If Badiscore's iOS rating count is still under ten by late December and its grossing rank has dropped out of the top 100, then "public track record" doesn't earn sustained payments even in Turkey, and trust isn't enough reason for users to pay. ② If in your chosen market the top 20 sports grossing apps each have hundreds or thousands of ratings, that market is no easier than Turkey. ③ Conversely, if Badiscore passes 100 ratings by year-end and holds a top 10 grossing spot in Turkey without big ad spend, then "verifiable" really does sell, and it's worth copying faster.
Data and sources
- Installs and charts: Google Play daily cumulative installs and the Turkey sports top free chart, 11-01 to 11-13; App Store sports grossing and top free charts for Turkey, Azerbaijan, and other stores, 10-28 to 11-13.
- Store pages: App Store Turkey (reviews, in-app prices ₺499.99 / ₺2,999.99, release notes; read 11-13). Google Play detail pages for Turkey and the US.
- Website: the Performance, How it works, How we choose, and Pro pages on badiscore.com and its sitemap, read 11-13. Domain registration dates come from public domain registration records; the first archived snapshot of the site comes from the Wayback Machine.
- Website traffic: Chrome User Experience Report (CrUX) bucket for October 2026.
- Developer: metekog.com and LinkedIn.
- Competitors: App Store and Google Play charts, ratings, and install counts, as of around Nov 13.
- Social and discussion: X, TikTok, and Instagram profiles; web and Reddit searches, mid-November.
- Anywhere marked "inference", "estimate", or "unverified" is an inference, an estimate, or something not verified.
Research notes: competitor table, acquisition evidence, full research report
- Competitors
Name Where Launched Scale How it differs from this one Akıllı Saha(Paulabs AI) iOS 2026-09-26 iOS Turkey sports grossing No. 4,51 by rating count; Play 10,837 installs; Turkey sports top free No. 32 A local Turkish AI prediction app from the same period and market, an order of magnitude bigger. Badiscore wins on its public track record and also offers basketball and a web version. TipsPro AI / BestTips(Venei Software) iOS 2024-09 iOS Turkey grossing No. 6,386 by rating count; Play 326k installs; IAP $4.99–79.99 An established "AI slip" subscription that sells recommendations. Badiscore sells probabilities plus a complete track record. Soccer Tips AI / Football Bet Tips - Scouter(Artlines) Android 2023-08 Play 1.12M installs; Turkey sports top free No. 30; iOS Turkey grossing No. 13 A multi-country tipster app with broad reach. Badiscore only does rule-based probabilities and has no ads. GoScore AI (MELODY YAZILIM, a Turkish company) Android 2021-12 Play 1.1M installs; IAP $3.99–149.99 A long-running local Turkish player, AI predictions; whether it publishes calibration is unverified. 11XScore AI Football(HRY Apps) Android 2025-08 Play 3.02M installs, +54k in the last 7 days, charted in 25 stores A heavily paid-acquisition-driven global AI predictions app; no Badiscore ad campaign found to date 180Score AI Soccer Predictions(4GROW) Android 2024-03 Play 2.93M installs; IAP $3.99–59.99 Nearly identical price range; 180Score is spreading across many countries Forebet / BetMines Android 2014 / 2019 Play 3.57M / 7.15M installs An established probability-prediction site and app with ads; Badiscore has no ads and uses rules to keep its three predictions consistent with each other FootyStats Web Around 2017 (App 2021-10) #21 Top Grossing on iOS Turkey; has Turkish-language prediction pages A site selling stats and probabilities; Badiscore is the newer three-platform product and also covers basketball Goal Signal / QuantGoal / Rega AI / Kahin AI iOS 2026-06 to 2026-11 Top Grossing on iOS Turkey at #16 / #17 / #22; ratings 51 / 8 / 11; Kahin AI has 1,362 Play installs A wave of Turkish indie-dev AI prediction apps of roughly Badiscore's size Mackolik, Sofascore, Flashscore, plus editorial picks from Nesine / Bilyoner Big-company features Since 2010 Play 38.1M / 108M / 106M installs; Nesine has 5.99M installs and is #2 in Turkey's free Sports chart Live-score giants and licensed betting platforms offer odds, polls and editorial picks for free; Badiscore has to rely on "verifiable probabilities" to get people to pay - Where users come from
- Verified Store search / ASO (Turkish keywords): The Play Turkey title is "Badiscore: Maç Tahminleri" ("match predictions"), with the description fully localized; iOS titles also vary by storefront ("Canlı Skor & Tahmin" in the NL store, "Fußball-Prognosen" in the CH store); rankings are mainly in Turkey and in Turkic-language Azerbaijan
- Verified Low-barrier category charts: Turkey's Play free Sports chart #46 needs only 1,024 installs; on iOS Turkey Sports Top Grossing, #22 and #23 Rega AI and Bowken have only 11 and 2 ratings
- Verified Web SEO (programmatic pages): The badiscore.com sitemap has 20,784 URLs (8 languages × team / matchup / league / basketball pages), and 2026-10's CrUX data shows it already in the top 500k in Turkey; the domain was registered only 2026-09-29 ago. Every page pushes the app download
- Inferred Suspected paid acquisition (around 11-10): The update notes for 1.8.7 (11-05) and 1.8.8 (11-07) added an "ad measurement" toggle; from 11-10 to 11-11, the iOS Turkey free chart went 123→21 and Top Grossing 135→16, Azerbaijan free 172→32, and the Play Turkey chart 123→46, with several stores jumping at once. Our Meta / Google / TikTok ad libraries don't track it, so no creatives can be found
- Verified Owned social (X / TikTok / Instagram): X @badiscore has only 8 posts; TikTok @badiscorecom has 7 followers and 21 videos, each getting 136–550 views; there's also Instagram @badiscore_us posting NBA breakdowns. Reach is too small to bring in new users
Badiscore: why it got surfaced, and whether it has actually broken out
Package name com.badiscore.app; iOS id 6819605253; web badiscore.com. The developer is Hasan Metehan Bacı (listed as metekog on Play), based in Ankara, with over 7 years of fintech and defense-simulation development (see metekog.com and LinkedIn). Research date: data as of 2026-11-13.
Conclusion first
It hasn't broken out yet. It has only surfaced in a very low-barrier market.
- On Play for 17 days with 1,024 cumulative installs; only 7 ratings on the iOS Turkey store.
- The impressive-looking ranks on the card come from several things stacking up:
- Turkey's Play free Sports chart #46 needs only 1,000 installs to reach;
- iOS Turkey Sports Top Grossing #16–30 needs only a few subscriptions to reach;
- The store title uses the high-frequency Turkish term "Maç Tahminleri" (match predictions);
- Turkish bettors were already willing to pay for "iddaa predictions."
- Its real differentiator is public track record, public prediction rules, and public probability calibration. That's what keeps users and earns good reviews, but it isn't what drives acquisition.
1. What it is
A football-plus-basketball product with pre-match probabilities and live scores; web, iOS and Android share one account, and the Pro subscription works across all three.
- Coverage: football covers 48 leagues across 35 countries; basketball includes the NBA, EuroLeague, EuroCup, and the domestic leagues of Spain, Italy, Germany, France, Greece and Turkey.
- Predictions per match: three for football (main pick, second choice, bold pick), one for basketball.
- Where the probabilities come from: goal and win/loss markets come from pre-match odds; corners, yellow cards, shots and saves use an in-house model (see How it works).
- Free vs. paid: scores and standings are free; registered members get 2 football matches plus 2 basketball matches per day (one of them an NBA game); Pro unlocks all matches, live probabilities, Streaks, and full stats.
- Pricing: Play lists $9.99–59.99 per item, monthly or annual (googleplay library iap_range; Pro page).
- Other: no ads, 18+.
Timeline
| Date | Event |
|---|---|
| 2025-12-29 | X account opened (date decoded from account id) |
| 2026-09-29 | Domain registered (RDAP) |
| 2026-09-30 | First captured by the Wayback Machine |
| 2026-10-10 | iOS release |
| 2026-10-27 | Listed on Play |
Between iOS 10-19 and 11-11 it shipped 1.3 → 1.8.11, roughly 10 versions, and Play also updated almost daily. Shipping this fast as one person likely means heavy use of AI coding.
2. Data
Play (googleplay library, app_day)
| Date | Total installs | Turkey sports free chart |
|---|---|---|
| 11-01 | 1 | 121 |
| 11-03 | 171 | 128 |
| 11-08 | 361 | 122 |
| 11-10 | 433 | 123 |
| 11-11 | 460 | 46 |
| 11-12 | 460 | 38 |
| 11-13 | 1,024 | 46 |
- Install counts jump in steps (+170, +167, +564), with days of 0 increments in between. That's Google updating the counter in batches, so don't trust single-day increments.
- It only charts in Turkey. However, the detail page loads from the US store, so it is not listed only in Turkey. The card's claim that it's "only listed in Turkey" is inaccurate.
- The Play Turkey page shows 4.6 stars, 9 reviews, and "1 B+" (1,000+) downloads.
iOS (appstore library, chart_rank, Turkey Sports category)
- Top Grossing: 10-28 #88,10-30 #54,11-03 #51,11-10 #135; 11-11 rose to #16, 11-12 #23,11-13 #30.
- Top Free: 11-06 #62,11-10 #37, 11-11 #21, 11-13 #74.
- Other stores: its best Azerbaijan Top Grossing rank was #8 (10-30), and Top Free #25 (11-11); Lithuania, Switzerland, Bulgaria and the Netherlands have also seen sporadic category-chart appearances.
- Ratings: 0 (10-31) → 7 (11-08), with no growth since. All 4 reviews are 5 stars, all in Turkish.
Web
- Tranco has no ranking for it.
- CrUX 2026-10 shows badiscore.com already in the top 500k in Turkey. In other words, a domain registered only a month ago is already getting real Chrome traffic (traffic library).
- The sitemap has 20,784 URLs: 8 languages (tr / en / de / es / fr / it / pt / el) × team, matchup, league and basketball pages, a classic programmatic SEO setup.
Benchmark: Turkey's Play free Sports #46 needs roughly 1,000 installs. On iOS Turkey Sports Top Grossing, #22 Rega AI has only 11 ratings and #23 Bowken only 2. So reaching "top 30" doesn't prove much on its own.
III. Competitors
The Turkish "maç tahmini / AI prediction" niche is very crowded. Among the top 80 iOS Turkey Sports Top Grossing apps, there are about 30 prediction apps, and at least 10 of them are work from Turkish indie devs that launched in the last 2026 years (appstore library app_profile).
| Competitors | Platform | Launched | Scale | How Badiscore is different |
|---|---|---|---|---|
| Akıllı Saha(Paulabs AI) | iOS / Android | 2026-09-26 | iOS Turkey Top Grossing #4,51 by rating; Play 10,837 installs, Turkey Sports Top Free #32 | Same period, same market, one order of magnitude larger |
| TipsPro AI / BestTips(Venei) | iOS / Android | 2024-09 | iOS Turkey Top Grossing #6,386 rating; Play 326k installs | Sells picks, not probabilities or track records |
| Scouter(Artlines) | Android / iOS | 2023 | Play 1.12M installs, #30 in Turkey's free chart; iOS Top Grossing #13 | Multi-country tipster spamming |
| GoScore AI (Turkish company MELODY) | Android / iOS | 2021-12 | 1.1M Play installs | Local veteran player |
| 11XScore AI Football | Android | 2025-08 | Play 3.02M installs, +54k in the last 7 days, charted in 25 stores | Heavy paid acquisition |
| 180Score | Android | 2024-03 | Play 2.93M installs; IAP $3.99–59.99 | Nearly identical pricing, global ad blanketing |
| Forebet / BetMines | Android / web | 2014 / 2019 | Play 3.57M / 7.15M installs | Long-established odds site with ads |
| FootyStats | Web / iOS | Around 2017 | iOS Turkey grossing No. 21 | Sells stats, with a local-slang prediction page |
| Goal Signal, QuantGoal (on 11-05), Rega AI, Kahin AI, Tahmin Makinesi, ForeKick, Temposcore, ScoreLab | iOS / Android | 2025-12 to 2026-11 | Ratings sit at a few to a few dozen; Kahin AI has 1,362 installs on Play and ranks #43 on the Turkey Sports free chart | Same wave, about the same size as Badiscore |
| Mackolik, Sofascore, Flashscore; editorial picks from Nesine, Bilyoner, Misli | Big-company features | Since 2010 | 38M to 100M installs; Nesine ranks #2 on the Turkey Sports free chart | Free odds, polls and editorial picks |
4. Where users come from
- Store search / ASO (confirmed)
- The Play Turkey title is "Badiscore: Maç Tahminleri", and the description is written entirely in Turkish (opening with Süper Lig and Türkiye Kupası).
- The iOS titles also differ by store: the nl store uses "Canlı Skor & Tahmin", the ch store uses "Fußball-Prognosen", and the US store uses "Soccer Predictions / Basketball & Soccer".
- The stores where it has a ranking are almost all Turkey and Azerbaijan, a Turkic-language market.
- Exposure from low-barrier category charts (confirmed): see the barrier comparison above.
- Web SEO (layout confirmed, effect not quantified): 20,784 programmatic pages in 8 languages; every page has a "Get the app" QR code, and downloads go through tracked links like
badiscore.com/app?k=web&to=ios; October has already reached the top 500k band in Turkey's CrUX data. - Suspected paid promotion (unconfirmed)
- The release notes for 1.8.7 (11-05) say "Privacy: manage analytics and ad measurement in one place", and 1.8.8 (11-07) says "if you decline ad measurement, your device ID is not shared", which suggests an ad attribution SDK is integrated.
- Then from 11-10 to 11-11, several storefronts (iOS Turkey, iOS Azerbaijan, Play Turkey) jumped at the same time. Those days coincided with international match days with few matches, so it doesn't look like organic match-driven traffic.
- Our Meta, Google and TikTok ad libraries don't track it, and we found no creatives, so no conclusion can be drawn.
- Owned social (confirmed, very small volume)
- X @badiscore: 8 posts, mostly three pre-match predictions and "minute by minute" recaps.
- TikTok @badiscorecom: 7 followers, 21 videos, 136–550 views each.
- Instagram @badiscore_us: does NBA breakdowns; the login wall blocks follower counts.
- No media or community: Jina, Exa and Brave turn up no coverage, Reddit posts or forum discussion; x-radar shows 0 mentions.
5. Why people stay and why it makes money
- The demand already exists: the top sports charts in Turkey are led by licensed iddaa platforms such as Nesine, iddaa.com, Bilyoner and Misli. Paying for "predictions / picks" is an established category locally. On iOS it reached the Turkey Sports Top Grossing chart on day 18 after launch (10-28), which shows some of the first users were paying.
- Solving the trust problem: the chronic issue in the tipster world is "deleting losing picks." An iOS review (2026-11-01) says, in Turkish: "Diğer tahmin uygulamaları gibi geçmiş tahminleri de silmiyorlar" (they don't delete past predictions like other prediction apps do). Other reviews praise "analizler çok iyi" (the analysis is very good) and the "seriler sayfası" (streak page).
- Public calibration data: the How it works page says that across the 703,127 options given a 60–65% probability, the actual hit rate is 62.5%; for basketball, 59,414 analyses hit at 61.5% against an expected 60.9%.
- Transparent rules: first set a "match expectation" (if the favorite's win rate is ≥ 45%, expect the favorite to win, otherwise look at total goals); then pick the options that hold together under that expectation at ≥ 85%; skip options above roughly two-thirds probability; skip odd/even, half-time and shot markets; re-pick only if the probability drops more than 8 points after lineups are announced.
- A reason to come back daily: each morning it selects 4 free matches, plus "My Day", "Streaks" and live scores.
- Connecting the web: subscriptions can be paid on the web (the site has a pre-information form and a distance sales agreement, documents required by Turkish e-commerce law), which avoids store commissions.
6. Differentiation, and whether it drove growth
Differentiation
- Verifiable: the track record is public and never deleted, there is calibration data, and the rules are published in writing.
- Structured three-pick output: one expectation yields a main, second and bold pick, and the three can hold at the same time.
- All three platforms in one, no ads, covering both football and basketball.
Is that what drove the growth? Mostly not.
- Acquisition came from the low barrier in the Turkey storefront and local keywords, plus possibly the 11-10 push.
- Differentiation affects conversion and word of mouth: one of the 4 5-star reviews explicitly says "it doesn't delete the track record."
- The probability core is pre-match odds plus a relatively simple in-house model; anyone could copy it in a week or two.
- Its moat is only the "public track record it has already built," and that takes time.
VII. Visibility and copycats
- Visibility: few people know about it yet. No major media, Reddit, HN or X accounts have written about it, and revenue is not public (not on TrustMRR, and the founders haven't posted it).
- Copycats: none specifically copying it, but in the same Turkish market at least ten similar indie AI prediction apps have appeared in 2026 years, so it's already a crowded niche.
8. What others can borrow
- Pick a small-language sports market with high willingness to pay and low chart barriers, and use the high-frequency local search terms directly in the title.
- Use "a public, undeletable track record + calibration data" as the trust signal, the thing the tipster category most lacks.
- Use programmatic web pages for SEO and the app for conversion, tie accounts and subscriptions together, and keep the money collected on the web without sharing it with the stores.
- Release a fixed set of free predictions every day to build a "come back and check every day" habit.
9. What I'm still unsure about
- The reason several storefronts charted at once, from 11-10 to 11-11: paid promotion, Telegram or influencer shares, or coincidence. None of it has been verified.
- Actual revenue and subscriber numbers are unknown, and Turkey-store TL prices weren't retrieved (the iTunes API was rate-limiting at the time).
- Play install counts are updated in batches, so single-day numbers are unreliable.
- How traffic splits between web and app is unclear; CrUX only gives a "top 500k band."
- Two points in the card need correcting:
- "Only app under this developer" holds only for Play. The iOS developer page (id1864635651) also lists Paw Rush and Ferpz AI.
- The "second listing with the same name" is actually the same id 6819605253, just with the title changed to "Badiscore: Basketball & Soccer" in the US store.
Sources
- Our libraries: googleplay (app_day, chart_rank, v_chart, app_profile, app), appstore (chart_rank, app_day, app_profile, app_snapshot / app_text, review, app_name), traffic (crux_domain), metaads / googleads / tiktokads (no records found); data as of 2026-11-13.
- Web: badiscore.com, Pro, How it works, How we choose, Play Turkey page, App Store Turkey page, X @badiscore, metekog.com, RDAP (Verisign), Wayback CDX, badiscore.com/sitemap.xml.
- TikTok data comes from TikHub (the @badiscorecom profile and search).