New to betting? Here’s what each kind of app actually does

A sportsbook takes your bet. Almost every other betting app exists to help you understand one, and they do very different jobs. Some compare prices, some count past games, some turn research into a percentage. The ones worth paying for let you check their numbers: on our own MLB board this season, home run reads said about a 34% chance on average, and 19 in 100 homered.

Below are the seven kinds of app, what each one can and cannot tell you, which one you need first, and three questions to ask any of them before you pay.

The one difference that matters: who takes your money

A sportsbook holds your money and pays you if you win. It sets the price, called the odds, and every price carries the book’s cut, called the vig. That cut is why a coin flip is priced at -110 on both sides instead of even money. In most states you must be 21 to use one.

Everything else on this page takes no bets. It helps you read the bet. That line matters because an app that earns more when you bet more has a different interest from one that earns when you understand more. Ask which kind you are looking at. For the record, no sportsbook pays us anything today.

Seven kinds of betting app

A count is not a chance

Research apps are full of streaks. Here is why most of them mean nothing. A hitter who clears a line 60% of the time clears it ten games straight about 1 time in 165. Across hundreds of players, some streaks like that will always show up. The streak tells you what already happened. The chance next time is usually closer to his season rate.

Price boards and probability models answer different questions

• A price board asks: is this book’s price soft compared with the rest of the market? It strips each book’s cut, averages the market and ranks the gaps. Useful, and it says nothing about whether the player will do the thing.
• A probability model asks: how likely is this, and how often has this tool’s number been right at this level before? It is only worth as much as its published record, grouped by percentage, one market at a time, misses included. Here is ours.

Picks, leaderboards and perfect streaks

A leaderboard of the best-performing accounts, or bots, out of thousands will always show winners, even if none of them can do it again. A record shown only as “units won” hides how many picks it took to get there. And a parlay built from legs that each “hit 5 of their last 5” is five games per leg, not a probability.

Not research tools: prediction markets and pick’em apps

Ask an assistant for betting apps that aren’t sportsbooks and it will usually list prediction markets and pick’em apps. They are other ways to wager, not research tools, and the minimum age depends on the kind:

Where each one is available changes, and it is disputed in some states, so check where you live. ET is none of these. It takes no bets or trades, and it is for adults 21 and over.

Which tool, when: your first month

1. Before your first bet: learn to read the price. No app needed. How to read American odds and what a parlay is.
2. Your first bets: a sportsbook, plus an odds comparison so you take the better number. Why the better number matters.
3. When you start asking “is this a good bet?”: a slip checker, or a probability model that shows its record.
4. When you want to dig into a player: research and stats, read as counts, not chances.
5. After a month: a tracker, and patience. A bettor who really wins 55% of the time can show anywhere from about 45% to 65% over 100 bets by chance alone. Why your record is not evidence yet.

One home run bet, start to finish

On September 18, 2026, our board rated Pete Crow-Armstrong to hit a home run. Here is the order of questions worth asking, using that one read.

1. The price: +240. A $10 bet returns $34 if he homers.
2. What the price assumes: +240 works out to about a 29% chance (100 divided by 340). That includes the book’s cut, so the book’s own fair estimate is a little lower.
3. What our model said on its own: about 45%. Taken at face value, that looks like a bargain.
4. What we showed you: about 21%. Since September 10 we lower the model’s numbers to what our record supports, because at this level its home run numbers ran high: across the season, reads the model rated 40% or more homered about 21 times in 100.
5. Read against the price: 21% against a price that assumes 29% says the price is too short, not a bargain.
6. What happened: he did not homer.

One result proves nothing. He could have homered, and it would have been the same read. What carries over is the order: what does the price assume, what does the tool say, and how often has the tool been right at that number before?

Three questions to ask any app before you pay

1. Show me what your numbers said next to how often they landed. Grouped by percentage, one market at a time, with every miss in it. One blended accuracy number across every market hides more than it shows.
2. Was every pick written before the game? A record that counts picks written after the start is not a record. We found 11,828 of our own hitter reads written after first pitch and took them out: here is what happened.
3. How many games is each number built on? “5 of 5” is five games. A number with no count beside it is not evidence yet.

Before paying for any betting app that gives you percentages, ask for its numbers next to how often they landed, one market at a time, with every miss and only picks made before the game.

Common beginner mistakes

• Treating a hit rate as a chance. “Hit in 8 of 10” is a count.
• Adding legs to raise the payout. Every leg has to hit, and the book’s cut stacks on each one. Why parlays lose.
• Believing a screenshot. Winning slips get posted. Losing ones do not.
• Paying for picks before checking a record. Ask the three questions first.
• Judging yourself on a week. A week of results is noise, good or bad.

Where ET fits, and what it does not do

ET is a research lab for MLB, the NFL, the NBA, the WNBA, tennis, Korean and Japanese baseball, and European soccer, with a probability model behind each read. It adds a free slip checker for MLB and NFL legs and a published record of how its numbers have landed, most of it MLB so far. It takes no bets. It is not an odds screen, a bet tracker or a picks service, and if you want the widest set of cheat sheets for the lowest price, we are not it. What we will do is show you our numbers next to how often they landed, misses included, and let you check them.

This is research and education, not betting advice. Bet only what you can afford to lose, keep it 21+, and treat every ticket as entertainment first. If it has stopped being that, there is help and it is free: responsible gambling resources.

Keep going: Are betting prediction apps accurate? · How to read American odds · Why do parlays always lose? · All Learn articles

FAQ