Ask most people running a mixed book what percentage of their money is in crypto and you get a guess. Not because they are careless, but because the holdings table is one long list and the eye cannot weight a list. Eleven rows read as eleven similar things whether the largest is 60 percent of the book or 6 percent.
The filter strip sitting above the table is the fix, and it is doing more work than a cosmetic sort. SPOT CRYPTO, PERP CRYPTO and STOCKS are the chips visible in the capture below, and since the platform also covers FX, commodities and prediction markets, a book holding those types carries the same problem across more markets than mine does. Those chips are not just filters, they mark populations that should never have been added together in the first place.
What the strip is actually separating
A spot crypto row is an asset you own. A stock row is an asset you own, settled through a completely different system with a completely different set of hours. A perp row is not an asset you own at all, it is an exposure carried on margin. An FX or prediction market position is different again in how it is funded and how it resolves.
The dollar figure printed next to each of those means something different in each case, which is exactly why the filters exist. Pressing one is the cheapest way to stop comparing quantities that are not comparable. It is also the only quick way to answer the allocation question honestly, because the answer changes depending on which population you put in the denominator, and you have to choose that denominator deliberately.
Every filter you press changes the denominator
This is the part that trips people up. Press SPOT CRYPTO and the table shows your spot crypto rows. Whatever total is now on screen is a total of that subset, not of your money. If you then read a row as 30 percent of the visible total, that is 30 percent of your spot crypto, which might be 12 percent of everything you own.
So before you press anything, write down the unfiltered total. In the capture that is 51,715.70 USD, and every share you calculate afterwards should be expressed against that number, not against whatever the filtered view shows. It sounds pedantic. It is the difference between believing you hold a 30 percent position and holding a 12 percent one, and those two beliefs produce different trades.
The routine that works is boring. Note the full total. Press one filter, add the rows in that view, write the subtotal. Press the next, do the same. When you are finished you have three or four subtotals that add back to the total, and you know your real allocation for the first time. It takes about ten minutes and most people are surprised by at least one of the numbers.
Perp rows break percent of book arithmetic
The one place where that adding up stops working cleanly is perps, and it is worth understanding before you include them in an allocation number.
A spot position of 5,000 dollars uses 5,000 dollars of your money. A perp position showing 5,000 dollars of exposure might be using 500 dollars of margin, or 1,000, depending on what you set. Those two rows can print similar looking figures while occupying completely different amounts of your capital and carrying completely different loss profiles. The spot position can fall 40 percent and still be a position. The perp position can be liquidated and stop existing.
What I do is keep two numbers rather than trying to force one. The first is capital allocation, which is what fraction of my money is committed where, and for perps that means the margin posted rather than the exposure shown. The second is exposure, which is what happens to me if crypto moves 10 percent, and for perps that means the full position size. A book that is 15 percent of capital in perps can easily be 60 percent of exposure to crypto, and the second number is the one that hurts you on a bad weekend.
If that sounds like extra work, the filter makes it fast. Press PERP CRYPTO, look at what is there, and if the answer is nothing then you can ignore this section entirely and treat your allocation percentages as clean.
An empty filter is a finding, not a blank screen
With 4 unique positions across 8 connected sources, most filters on that account are going to come back empty or nearly empty. That is not a failure of the view, it is the most useful thing it can tell you.
Two conclusions come out of an empty filter and they need different responses. The first is that you genuinely have nothing in that market, in which case the question is whether you meant to. Plenty of people describe themselves as diversified across crypto and equities while holding, in practice, four crypto positions and one of them a stablecoin. The filters settle that argument in five seconds.
The second conclusion is that you do hold something in that market but the venue holding it is not connected. That is a different problem and a more dangerous one, because the total you have been sizing against is missing a chunk of your money. Check the ACCOUNTS tile against the list of places you actually have money. On the capture it reads 8 connected sources, and if your real answer is nine or ten, the missing ones are the first thing to fix before any of these percentages mean anything.
The pass I run before adding a position
The point of all this is not tidiness, it is that the next trade you place should be sized against a real picture rather than a vibe. Before I add anything to a mixed book I run this:
- Read the unfiltered total and write it at the top of a page. That is the denominator for everything else.
- Press each type filter in turn and record the subtotal. Confirm the subtotals add back to the total. If they do not, something is in a category you did not expect and it is worth finding out what.
- Convert each subtotal to a percentage of the full total. Compare that list to what you would have guessed before you started. The gap between the two is the value of the exercise.
- For perps, replace exposure with margin posted for the capital number, and keep the exposure figure separately for the risk number.
- Ask what the position you are about to add does to the largest of those percentages. If it pushes the biggest market past the share you decided you were comfortable with, the trade is either smaller than you planned or funded by trimming something else.
The last step is the one that changes behaviour. Almost nobody adds a position by deciding to increase their crypto weight from 68 to 74 percent, and almost everybody does exactly that by accident, one reasonable looking trade at a time. The filters are how you catch it before the trade rather than during the drawdown.