A betting bot and a person betting by hand place the same bet. What differs is how many, how fast, and how consistently, so automation changes the volume of your betting rather than the quality of it. If your rule has an edge, automation applies that edge more often. If it does not, automation reaches the losses sooner.
That is the decision in one paragraph. The rest of this page is the arithmetic behind it: what a bot does better than a person, what it does not do at all, what it costs, and the test for whether your rule is ready.
What automation actually changes
Three things decide what a betting rule returns: the edge on each bet, the stake, and the number of bets. Automation moves one of them.
A person is limited by attention. You have to be at the screen when the price is there, and you will not apply the same rule at 11pm on a Tuesday as you did at 2pm on Saturday. A bot has no opinion about the time of day: it checks the condition, and where the condition is met it places the bet. What it does not move is the edge, which comes from the rule, which is to say from which price, measured against what fair price, in which market.
| What decides the result | Moved by automation |
|---|---|
| Edge per bet | No, it comes from your rule and the price you get |
| Stake per bet | No, it applies the staking plan you set |
| Number of bets | Yes, by a lot |
| Consistency of application | Yes |
What a bot does better than a person
Four things, and all four are mechanical rather than clever.
- Being there at the price. Racing prices can move sharply in the last few minutes before the jump, and software checking every few seconds acts on a number that is gone before a person has finished reading the screen.
- Holding a session at each of your bookmakers. A person logs into one bookmaker at a time, while software can hold a logged-in session per account and act in each when the condition is met. Every account has to be one you hold in your own name, and many bookmakers restrict automated betting, third-party access and multiple accounts in their terms.
- Applying the rule the same way every time. No tilt, no "this one looks good", no quietly skipping the selection your rule actually picked.
- Recording every bet. Every placement, price and result in one place, which is what lets you tell whether the rule is working.
What a bot does not do
This is the half of the comparison a sales pitch leaves out.
- It does not find an edge. Which price is worth taking is answered before the software ever runs.
- It does not pick selections or rate runners. A price comparison rule is not a form analysis.
- It does not change what a bookmaker's terms allow. Bookmaker terms govern the account rather than the software, and many bookmakers restrict or prohibit automated betting, third-party access and multiple accounts in those terms. Under them a bookmaker can refuse a bet, void a bet, cap the stake it will accept, or close the account.
- It does not keep betting once an account is limited or closed, and no software prevents an account being restricted or closed. A rule that needs $200 a bet stops being worth running the day that account is capped at $5. See why bookmakers restrict accounts.
- It does not work while the computer is off, asleep or offline.
The same rule by hand and automated
Illustrative figures, chosen to show the arithmetic rather than to describe anyone's results.
One rule: back a selection where the price you can get is 2% above your fair price, at a flat $20 a bet. The expected value of one such bet is stake x edge, so 20 x 0.02 = $0.40. By hand you get to 15 of those a week. Automated across your own accounts at several bookmakers, the same rule finds 90, six times as many, and the edge does not change: still 2%, still 40 cents a bet. Point the edge the other way, which is what a stale fair price or a wrong rule gives you, and each bet is worth minus 40 cents instead.
| By hand | Automated | |
|---|---|---|
| Bets a week | 15 | 90 |
| Turnover a week | 15 x 20 = $300 | 90 x 20 = $1,800 |
| Expected a week at a 2% edge | 15 x 0.40 = $6.00 | 90 x 0.40 = $36.00 |
| Expected over 12 weeks | 12 x 6 = $72 | 12 x 36 = $432 |
| Expected a week at a minus 2% edge | minus $6.00 | minus $36.00 |
| And over 12 weeks | minus $72 | minus $432 |
Six times the bets is six times the expected result in whichever direction the edge actually points. That is why the first question is never "should I automate" but "is the edge real".
Risk: Betting involves risk. Bookmakers can restrict or close accounts and void bets, automation can fail, prices move, and a positive expected value bet can still lose. There is no guarantee of profit. See responsible gambling for limits and support. Many bookmakers restrict automated betting in their terms, and automating a rule with no edge only reaches the losses faster.
Volume cuts both ways, which is the responsible gambling point
Automation makes it possible to bet far more, far faster, than by hand. In the table above that was the upside. It is the same fact as the risk, and it is the reason to set your limits before you automate rather than after.
The firmest cap is the one a bookmaker enforces. A deposit limit is set with each bookmaker, not inside any software, and a limit at one bookmaker does not apply at another, so a rule running across several accounts means setting a limit at each. Software stops and caps, such as run windows, stake maximums and loss cut-offs, are ordinary settings rather than guarantees: they can fail or act late. If betting has stopped being something you control, the services on responsible gambling are the place to start, and BetStop explains the national self-exclusion register.
What each route costs
A bot against hand comparison usually stops at "it saves time". The two routes do not carry the same kind of cost, and that difference decides more of this than the time does.
Betting by hand costs your attention and nothing else. A quiet week costs nothing, because there is nothing to pay for a rule that never fired.
Automating adds three costs that do not care whether the rule fired: a plan price, charged whether you bet or not; per-bet usage charges, which grow with every bet placed; and a computer that has to stay on, awake and online while a session runs. Your attention does not fall to zero either. Someone still reconciles the records, notices a bookmaker refusing bets, and stops a session behaving oddly.
So a plan price is a volume question rather than a price question. Divide the monthly cost by the turnover you expect in a month, and the answer is the edge you have to clear before the tool has paid for itself: at a few bets a week that number is far above what a price comparison realistically finds, and it falls only as the bets rise. Do betting bots make money works that arithmetic through with a table, and the betting software buyers checklist has the charges to get in writing first.
The decision rule: write the rule down first
Write your rule as something a stranger could follow without asking you a single question: which market, which price, measured against what, at what stake, inside what window, and when not to bet.
If you can write it, automation is a fair tool for it, because a bot is precisely a rule applied without discretion. If you cannot, automation will not supply the missing part. It will apply your vagueness at speed and leave you with several hundred bets whose record no longer tells you which judgement produced them. The common version is a rule that is really "a price that looks good to me", which is not a rule at all.
Two steps beat guessing. Run the rule by hand for a few weeks and keep the record, because if it does not survive a small sample applied by a person, volume will not rescue it. Then measure the price you took against the market's last price before the off, which says whether you are beating the market long before profit does. See closing line value.
When betting by hand is the better choice
Automation is not the default answer. Betting by hand is the better choice when any of these is true.
- You have no written rule yet, only a feel for which price looks worth taking.
- Your edge comes from judgement software cannot hold: a reading of the form, a late look at the mounting yard, an opinion about a team.
- The volume is low. A rule that fires twice a week never spreads a fixed plan price.
- You want to think about each bet, which is a different activity from running a rule at volume.
- You are not willing to hand bookmaker logins to software. See giving betting software your login.
The two are also not exclusive, and a common settled position is a written rule running automated with the judgement bets still placed by hand. What breaks is the record: two routes into one bankroll, where a bet you placed yourself does not appear in the software's record at all unless you add it. Keep the two labelled, so you can still tell which route produced which result. If you cannot, you have no way of knowing whether automating was the part that helped.
Where B337 fits
B337 is betting automation and odds comparison software for adults in Australia. It runs on your own computer and places bets only in bookmaker and exchange accounts you hold, on the rules you set. It is software rather than a bookmaker: it does not take bets, set prices or hold your betting funds, which stay in your own accounts under each bookmaker's terms.
On the manual side of this comparison, Terminal View is read-only access to the odds board at $25 a month, and Terminal Pro at $300 a month adds placing bets by hand from the board into your own accounts. On the automated side, the strategies that run sessions come with Full Automation, set up with the team and with no published price, and bets placed through B337 use credits, a per-bet usage charge on top of the plan price. For free and confidential support call 1800 858 858 or visit gamblinghelponline.org.au.
The limits are the half this page is about. Strategies follow the rules you set: B337 does not pick bets with a model, and the racing strategy compares prices rather than rating runners. It cannot keep betting once a bookmaker limits or closes an account, it cannot change what a bookmaker's terms allow, and it makes no claim that it avoids limits. Nothing is placed while your computer is off, and every placement, price and settlement is recorded in your dashboard, so each one can be checked against your bookmaker account.
Risk: Betting involves risk. Many bookmakers restrict automated betting in their terms, and no software prevents an account being restricted or closed. There is no guarantee of profit. See responsible gambling for limits and support.
For the mechanics, see how betting bots work, and for the failure modes and the controls that limit each one, see betting bot risks.