The Same Thing in Two Places
Every round trip in this course since lesson 63 has cost 0.1230 a share, and no page has ever opened it. Open it: a penny of spread, five basis points of slippage a side on a price near 103, and half a cent of commission, which is 0.0100 and 0.1030 and 0.0100, and which comes to 11.94 basis points. A second place to trade the same instrument competes for the first and the third of those. It cannot touch the second, which is 83.7 per cent of the bill, because slippage is lesson 69’s delay and a delay does not care which venue the order is sent to. So the whole prize a second venue offers is 0.0200 a share, and any price difference between the two has to clear 11.94 basis points before you keep a single cent of it.
Prerequisites: Lesson 63, for the 0.1230 and the three numbers it is made of, lesson 82, for the divisor and the list this page adds to, and lesson 69, for what a delay costs.
Taking the round trip apart
Lesson 63 stated its costs in one sentence and then spent the rest of the course charging them. Rebuild the sentence and the three parts do not divide the bill evenly, which is the whole of this page.
| Part of the round trip | A share | Basis points at 103 | Share of the bill | Does a second venue reach it? |
|---|---|---|---|---|
| Spread, a penny | 0.0100 | 0.97 | 8.1% | Yes |
| Slippage, five basis points a side | 0.1030 | 10.00 | 83.7% | No |
| Commission, half a cent a side | 0.0100 | 0.97 | 8.1% | Yes |
| The whole round trip | 0.1230 | 11.94 | 100% | A sixth of it |
The three add to 0.1230 exactly, which is the figure lesson 63 printed, so the bill reproduces from its own parts. Now read the last column. The spread is what the venue charges you for immediacy and the commission is what it charges you for access, and both of those are things a second venue can undercut. Slippage is the distance the price travels between the moment you decide and the moment you are filled, and lesson 69 measured that distance as a function of how long you take. Opening an account somewhere else does not make you faster.
So a second venue is bidding for 0.0200 of a 0.1230 bill, which is 16.3 per cent, and it is bidding against whatever your first venue already charges rather than against zero. The part that is out of its reach is five times larger than the part that is in it.
Almost nobody has split their own round trip into these three parts. It is one line on a broker statement against one line on a fill report, and it tells you which of your costs a different account could ever have changed.
Module 11’s list of the degrees of freedom this course did not have gains its second item: a second place to trade the same thing.
There is the other reason people open the second account, which is not to pay less but to trade the difference. Buy where it is cheap and sell where it is dear, and what you keep is the difference less one round trip, because you pay to enter on one side and to exit on the other exactly as you would in one place.
| Difference between the two venues | A share | Basis points | Left after the round trip | Share of the difference you keep |
|---|---|---|---|---|
| Half a round trip | 0.0615 | 5.97 | −0.0615 | None |
| One round trip | 0.1230 | 11.94 | 0.0000 | None |
| One and a half | 0.1845 | 17.91 | 0.0615 | 33.3% |
| Twice | 0.2460 | 23.88 | 0.1230 | 50.0% |
| Three times | 0.3690 | 35.83 | 0.2460 | 66.7% |
| Five times | 0.6150 | 59.71 | 0.4920 | 80.0% |
The second row is the one that decides whether any of this is a business. A difference of a whole round trip is worth nothing, and the difference has to reach one and a half round trips before you keep a third of it and twice before you keep half. Twelve basis points is the floor, twenty-four is where it starts to look like a trade, and every screenshot of two venues quoting different prices is being compared against zero rather than against twelve.
And put lesson 82’s divisor on it before anything else, because that settles what a second venue is. Two positions in the same instrument correlate at one, and the divisor returns two divided by one plus one, which is exactly 1.00 bets. A second instrument at least bought 0.04 of a bet. A second venue buys none.
What the second account would have paid for
Run it on the only record this course has. Lesson 63’s winning cell took seven round trips, and seven times 0.1230 is 0.8610, against the 0.86 a share the page printed as the gap between its gross 10.70 and its net 9.84. The cost line reproduces to the cent.
Net edge over holding: 4.16 a share. Gross edge over holding: 5.02.
Costs are therefore 17.1 per cent of the gross edge, and removing all of them would multiply the edge by 1.21.
A second venue reaches 0.0200 of each round trip, which over seven of them is 0.14 a share.
That lifts the edge over holding from 4.16 to 4.30, a factor of 1.03.
Now hold that against one thing the course has already priced. Lesson 69 measured five minutes between the signal and the order at 32.10 basis points in the small cap. The second venue’s entire gift across the whole seven-trade record is 0.14 a share, which on a price near 103 is 13.59 basis points. Being five minutes late once, on the instrument lesson 69 measured, costs more than a second account gives back across every trade in the record.
A second venue does not give you another bet. It gives you back part of one bill.
Which puts the two accounts in the right order. The account you open second is worth a sixth of a bill you have never split, and the minutes between your alert and your order are worth multiples of the whole thing. If you are going to spend an evening on your execution, spend it on the delay, and open the second account when you have a measured difference that clears twelve basis points often enough to be worth the paperwork.
So split your own round trip into spread, slippage and commission before you open anything anywhere.
What this does not settle
That this is about one asset class. It is about one instrument quoted in two places, which is a market structure rather than a thing you trade: a share on two exchanges, a future against the cash it settles to and a coin on two venues are the same arithmetic. What a market that never closes adds is that lesson 69’s delay has no close to hide behind, so the five minutes are five minutes at four in the morning as well.
That the sixth is either worth having or worth ignoring. It is worth having: a sixth of a bill that lesson 80 showed comes straight out of the edge, and on lesson 63’s own record fourteen cents a share. It is also not much: removing the entire round trip would multiply the edge over holding by 1.21 and the sixth a venue can reach multiplies it by 1.03. Both of those are true and the page prints both.
That two prices are two bets. Lesson 82’s divisor answers that in one line and the answer is 1.00, because the two positions correlate at one by construction. Everything a second venue can do for you is on the cost side of the arithmetic, and a page that treated it as diversification would be making the mistake module 9 spent five lessons on.
That lesson 69’s 32.10 basis points belongs on this instrument. It does not. That figure is the small cap and this page’s 11.94 is an instrument near 103, so the comparison is between two magnitudes and not between two measurements of the same market. What survives the objection is the ordering rather than the ratio: everywhere this course has measured a delay it has come out in tens of basis points, and a cross-venue difference has to clear about twelve before it is worth anything.
And the concession that costs most: this course has never measured a second venue. Every figure on this page is lesson 63’s own cost sentence taken apart and put back together, and the one input the arithmetic actually needs — how far apart two real venues quote the same thing, and for how long — is not here and cannot be, because the whole record is sixty daily closes of one market that closes. So the page can tell you what a second price is worth and it cannot tell you what never closing is worth. Module 11 has one more missing degree of freedom to name, and it is the one every page in this course has assumed away: the person who has to sit through all of it. Lesson 65 already priced one of their habits, at 24.25 per cent against 5.05, and lesson 84 asks what the rest of them cost.
Problems
- Split your own round trip. Take one recent trade and separate what you paid into three: the quoted spread you crossed, the difference between the price when you decided and the price you were filled at, and the commission. Ten minutes, and you end holding one number, the slippage share of your own bill, which is the part no second account can reach.
- Price your own floor in basis points. Divide your whole round trip by the price you paid and multiply by ten thousand. Half an hour, and you end holding one number, the difference between two venues that leaves you exactly nothing, which you compare against the 11.94 on this page before you believe any screenshot.
- Count how often the difference clears it. Take two venues you can actually reach, record the price of the same instrument on both once an hour for a week, and count the readings where the gap exceeds the floor from problem two. An evening spread over seven days, and you end holding one number, the share of hours in which the trade exists at all, which is the base rate every story about this trade leaves out.
Sources. Lawrence Harris, Trading and Exchanges: Market Microstructure for Practitioners (Oxford University Press, 2003), for where each part of a round trip comes from and for why the three parts answer to different things, which is the first table’s last column. Maureen O’Hara and Mao Ye, “Is Market Fragmentation Harming Market Quality?” (Journal of Financial Economics, 2011), for what actually happens to spreads when one instrument trades in many places, measured rather than assumed. Igor Makarov and Antoinette Schoar, “Trading and Arbitrage in Cryptocurrency Markets” (Journal of Financial Economics, 2020), for the measurement this page does not have: how far apart venues quote the same coin and how long the gap lasts. Eric Budish, Peter Cramton and John Shim, “The High-Frequency Trading Arms Race: Frequent Batch Auctions as a Market Design Response” (Quarterly Journal of Economics, 2015), for the finding that the size of a cross-venue difference does not shrink as everyone gets faster, only the time you have to take it.
Backtesting as Evidence
The 0.1230 a share this page takes apart, and the seven round trips it reproduces.
Read Lesson →The Delay You Remove
The 32.10 basis points that a second account cannot compete with.
Read Lesson →The Second Way to Disagree
The divisor that gives a second venue exactly 1.00 bets.
Read Lesson →Educational only. Trading involves substantial risk of loss. Not financial advice. Past performance does not guarantee future results.
💬 Discussion (0 comments)
Loading comments...
Ready to Trade with Signal Pilot?
Apply your trading education with professional indicators and real-time market analysis tools.
Back to Signal Pilot →