Listening Lab / 02
Your songs.
Your order. Your reasons.
Build a personal ranking from a journal shortlist or your own tracks. Compare two at a time, adjust the order, and keep a reason beside every close decision.
Start with a shortlist
Presets contain titles from the journal and start in alphabetical order, not in the editor's ranking order. They are starting sets, not a claim to include every version.
Your entries stay in your browser.
Your ranking
Number 1 is your highest placement. Use the arrow buttons to move a track, then write the reason you prefer it to its neighbours.
No-preference choices place A next to B, after B. The result still has unique positions rather than shared ranks. Finish or stop this step before exporting.
Keep your result
One saved ranking draft per browser. Export JSON to keep multiple lists or move your result to another device.
How paired sorting works
The first song becomes a one-item ordered list. Each following song is compared with a song near the middle of that list. Choosing A or B narrows the available insertion position until the new song can be placed. The process repeats until every song is in the list.
This is binary insertion driven by your decisions, not a tournament score. It avoids asking for every possible pair, but it assumes that your preferences can be represented by one consistent order. Choosing a different starting order or using “no preference” can affect the result, especially when songs are closely matched.
A three-song example
Imagine three fictional tracks: A, B and C. You prefer A to B, so A sits higher. If C belongs below A but above B, the resulting order is A, C, B. No track has gained an objective quality score; the list only records the preferences entered during those comparisons.
If you prefer A to B, B to C and C to A, those choices form a cycle. A single ordered list cannot express all three at once. This tool does not test every pair for cycles. Use the final manual check to revise the priority, separate the songs into different lists, or keep an acknowledged close call.
Make the last pass more useful than the first
Inspect neighbouring songs and write why the higher one wins under your stated priority. “I like it more” records a result but not a reason. “Its chorus develops the opening idea, while the next track mostly repeats its hook” gives you something specific to revisit.
Try a second list with a different goal only when that comparison is useful. A game soundtrack can be ranked for its role during play and separately for standalone listening. Do not combine those two lists into an invented average and present it as a public consensus.
Storage, imports and limits
Will this change the published journal ranking?
No. Your work is private, and there is no submission or voting endpoint. The source article remains unchanged. Export your result to share it yourself.
Why are there no live rankings or vote totals?
The site does not collect votes for this tool. Showing a crowd score without that data would misrepresent a private sorting exercise. All results come from the list and decisions you enter.
Which export should I keep?
Text is convenient for reading, CSV for a spreadsheet, and JSON for restoring the editable list. CSV cells that might be interpreted as formulas are neutralised with a leading apostrophe. Print creates a clean list with your reasons.
Why not hundreds of tracks?
The workspace is limited to forty titles so that comparisons and written reasons remain manageable. Split a large catalog by release, period or purpose. More comparisons do not make a subjective result scientifically precise.