# Merge 1: twenty strategy and measurement results

Cash planning is the leading candidate in these supplied results: `combine-cash` has the highest absolute holdout, with a positive paired comparison recorded by its author. This is a local candidate ranking, not a verified move to first on Paradigm; no new simulations, external source verification, or public submission were performed for this merge. [S9]

All twenty input files were present and parseable, and all have `blocked: null`. The measurement-only items true-mean, oracle-bound, draw-variance and ablation-map returned the base unchanged; draw-variance nevertheless left its larger measurement plan incomplete. Claims below are attributed reports from these artifacts, not independently reproduced experiments. [S1], [S2], [S3], [S4]

The table copies the top-level numeric fields exactly, including differing base reproductions. Base means refer to seeds 0–599; evaluation total/retail/arb are seeds 1000–1599; every table holdout is 600 simulations starting at 30000, and every top-level evaluation and holdout reports zero failures. Arbitrage edge is signed (negative means loss); Python is the item-reported version, and code bytes are computed as the UTF-8 length of its decoded `code` string, excluding JSON escaping. Lane labels group the stated directions; no strategy code is reproduced. [S1], [S2], [S3], [S4], [S5], [S6], [S7], [S8], [S9], [S10], [S11], [S12], [S13], [S14], [S15], [S16], [S17], [S18], [S19], [S20]

| Key / source | Lane / direction | Reproduced base | 1000–1599 total | Retail | Arbitrage | Holdout total | Python | Code bytes |
|---|---|---:|---:|---:|---:|---:|---|---:|
| true-mean [S1] | Measure / independent true mean | 102.33197908899267 | 107.2442092237515 | 135.80165652288784 | -28.557447299150834 | 96.0405203261775 | 3.12.3 | 13882 |
| oracle-bound [S2] | Measure / hidden-information value | 102.33197908899267 | 107.2442092237515 | 135.80165652288784 | -28.557447299150834 | 96.0405203261775 | 3.12.3 | 13882 |
| draw-variance [S3] | Measure / draw variance | 102.33197908899267 | 107.2442092237515 | 135.80165652288784 | -28.557447299150834 | 96.0405203261775 | 3.12.3 | 13882 |
| ablation-map [S4] | Remove & measure / component ablations | 102.33197908899267 | 107.2442092237515 | 135.80165652288784 | -28.557447299150834 | 96.0405203261775 | 3.13.5 | 13882 |
| leaders-gap [S5] | Measure & modify / stock-overpayment gate | 102.33197908899267 | 108.02413592598784 | 131.2171054050378 | -23.192969479067667 | 95.524825093741 | 3.12.3 | 13988 |
| combine-belief [S6] | Combine / latent belief | 102.331987422326 | 107.7287098973095 | 134.15892612772782 | -26.430216230414665 | 94.98058276501934 | 3.14.4 | 17022 |
| combine-overflow [S7] | Combine / overflow and FIFO slices | 102.33197908899267 | 108.42885016453633 | 136.803363239503 | -28.37451307498033 | 96.32372908023733 | 3.12.3 | 16350 |
| combine-two-cent [S8] | Combine / scarce-stock two-cent queue | 102.3319790889927 | 107.47676082579723 | 135.99677430634952 | -28.520013480566334 | 96.03782167039321 | 3.10.12 | 16523 |
| combine-cash [S9] | Combine / cover bank and cash reserve | 102.33197908899267 | 109.27254289017199 | 135.356137274445 | -26.083594384290333 | 98.12413933736983 | 3.12.3 | 14588 |
| combine-constants [S10] | Combine / transfer constants | 102.33197908899267 | 107.45204715775166 | 135.97942233023198 | -28.52737517249417 | 96.458133898515 | 3.12.3 | 13882 |
| combine-late-game [S11] | Combine / late-game stock rules | 102.33197908899267 | 109.29650163975117 | 134.38229969436017 | -25.085798054621666 | 97.4026568625055 | 3.12.3 | 13925 |
| combine-pair [S12] | Combine / belief plus overflow | 102.33197908899267 | 107.77941648794467 | 137.01055490399065 | -29.231138416059004 | 96.0368042128295 | 3.12.3 | 22855 |
| combine-three [S13] | Combine / belief, cash, two-cent | 102.331987422326 | 108.54670913062401 | 136.602853288088 | -28.056144157473664 | 96.739743692969 | 3.14.4 | 20236 |
| combine-all [S14] | Combine / sequential six-way graft | 102.331987422326 | 108.82311690588183 | 137.79040742643517 | -28.967290520565 | 95.6841292191785 | 3.14.4 | 15867 |
| combine-vote [S15] | Combine / state-dependent adviser vote | 102.331987422326 | 108.21374422633767 | 136.147655910821 | -27.933911684499666 | 96.09686103077634 | 3.14.4 | 30537 |
| demand-fit [S16] | Deepen / censored retail-demand fit | 102.331987422326 | 109.14949031033433 | 134.9172763842037 | -25.767786073874166 | 97.23073075564217 | 3.14.4 | 37608 |
| tick1-jump-cost [S17] | Deepen / tick-1 jump exposure | 102.331987422326 | 107.83578651840733 | 135.61937036092417 | -27.78358384253083 | 95.50209429163283 | 3.14.4 | 14401 |
| tick2-queue [S18] | Deepen / tick-2 FIFO commitment | 102.33197908899267 | 108.30384238668634 | 136.78007252256066 | -28.476230135887167 | 96.33203752220284 | 3.12.3 | 14387 |
| mid-jumps [S19] | Deepen / asymmetric mid-price jump loss | 102.33197908899267 | 108.00451371283867 | 135.11283915310833 | -27.108325440305002 | 96.19078601230667 | 3.15.0 | 14745 |
| near-tick [S20] | Deepen / spread-1 near-tick rules | 102.33197908899267 | 108.408889148626 | 135.82225180532035 | -27.413362656706333 | 96.77537839876851 | 3.12.3 | 14743 |

## Item findings

**true-mean.** Returned the base unchanged and measured an independent reference on seeds 40000-43999: mean 99.23869339150623, retail 127.07987834308194, arbitrage -27.8411849515945, with 4000 simulations and zero failures. Its top-decile share is 0.5265243835574145 and estimated standard error of a 200-run mean is 12.37313011835025, making small unpaired leaderboard gaps weak evidence. Next, keep that reference fixed for paired comparisons and freeze candidates before a different final holdout. [S1]

**oracle-bound.** Returned the base unchanged; scratch oracle experiments on 1000-1599 gave +1.05 for retail parameters, +0.33 for jump parameters, +1.93 for current probability, +8.78 for the next jump flag, +15.83 for current probability plus jump flag, +24.34 for next-step probability, and +34.10 for everything together (the detailed tried entries' precision). These are hidden-information interventions in this particular policy, without oracle holdout results; they suggest one-step quote exposure matters, but are neither attainable strategy gains nor rigorous global upper bounds on alternative policies. Next, examine observable signals or state-dependent exposure, while respecting the item's warning that Poisson jumps are memoryless and its parameter-estimation verdict need not apply to a different policy. [S2]

**draw-variance.** Returned the base unchanged, but the planned 30 x 200 measurement was incomplete: it reports 3200 simulations for feature analysis and 14 x 200 block means ranging 77.5 to 112.6, with mean 96.9 and sd 10.5. Settlement dominates its diagnostics (NO 173.2 versus YES 18.0), top 20 of 200 carry 51.6% of edge, and actual jump counts were unavailable in the simulator JSON, so jump intensity was only a proxy. Next, compare paired common-seed differences, examine settlement/spread strata without silently changing the target distribution, and instrument actual jumps. [S3]

**ablation-map.** Returned the base unchanged after independent component removals: base-minus-disabled worth was 131.861019497087 for probes, 10.569825371544 for the sub-tick filter, 14.2184466903835 for the large tick-1 bid, 11.838959767283 for order retention, 8.569734166729667 for spread-1 rules, 0.02886033333333323 for the last-step rule, and -0.986319175898833 for other YES ammunition. The last two intervals include zero; joint stock deletion costs 24.3924403209095 and the spread-1 decompositions overlap, so conditional marginal gains cannot be summed. Next, independently test removing only the other ammunition while preserving the large tick-1 bid; inactive pre/nb1/gs cleanup is behaviorally neutral. [S4]

**leaders-gap.** Added an inferred-lower-bound gate on deliberate low-price stock overpayment while retaining the large tick-1 bid; on the 100-seed diagnostic, planned pre-step loss fell from -8.875961181735567 to -2.641936319512476. Its paired holdout change is -0.5156952324365005 with standard error 0.892493448909843: arbitrage improves but lost retail prevents a demonstrated improvement, and the task flags one external source as unreliable, so neither its external leaderboard framing nor a causal generalization is independently established here. Next, measure the foregone retail sales and use a public-state inventory-demand rule rather than blanket stock caps. [S5]

**combine-belief.** Replaced the grid posterior and manual sub-cent decay with a censored latent-score mixture filter, selecting assumed jump sigma=0.4 and posterior sizing weight=2.0. Tuning paired gain 0.48450067355800286 has standard error 1.045146739766478, while holdout paired gain is -1.0599375611583355 with standard error 0.49740626976596103; one external source is flagged, so the filter's provenance is an item claim, while these are its reported local measurements. Next, keep the base pending fresh evidence and run the latent filter in shadow, testing hard probe bounds and price-band loss calibration. [S6]

**combine-overflow.** Split main quotes into four FIFO slices and added cash-capped competitor-price overflow on spreads above one tick, retaining the base estimator and stock mechanism. Paired tuning gain is 1.1846409407848328 with standard error 0.31792630781541154, but holdout gain is 0.2832087540598333 with standard error 0.3651905827107454, and direct overflow fill edge was negative despite the total-policy improvement. Next, reserve the same modeled cash without placing overflow to distinguish cash throttling from retail-burst capture, then validate on fresh seeds. [S7]

**combine-two-cent.** Kept one-cent sales as default, enabling a retained two-cent queue only under estimated stock scarcity, plus a small stock-backed overflow ask; broad one-cent replacement lost on tuning. Its own holdout base is 96.00139032710622, yielding paired gain 0.036431343287000005 with standard error 0.01989893793084752, rather than a comparison against the other items' 96.0405203261775; only 7 tuning simulations changed, and one external source is flagged, so the claimed graft and tiny lift deserve replication. Next, correct survival-conditioned demand estimates for censoring and replicate before further threshold searches. [S8]

**combine-cash.** Added a 50-step immediate-cover target, a separate 2000-step cheap crossing bank, a $200 reserve released over the final 50 steps, a 2.0-cent premium-cover cutoff, and a low-inventory ask charge up to 0.0025. Paired tuning gain is 2.0283336664204996 with standard error 0.35262349203866866 and paired holdout gain is 2.083619011192334 with standard error 0.4067546948632805, the strongest reported absolute candidate holdout in this group. Next, estimate retail flow per run to size the contingency bank and test on a new tuning range and untouched holdout. [S9]

**combine-constants.** Transferred only kLb 0.08 -> 0.04 after testing the earlier constants separately on two tuning halves; five of six failed to transfer and all six together lost 1.87 a run. The item reports tuning +0.21±0.08 and holdout +0.42±0.25, supporting at most a modest effect rather than its stronger phrase 'small and real' as a settled true-mean conclusion. Next, test regime-conditional sizing and demand-dependent YES targets instead of another global sweep. [S10]

**combine-late-game.** Rejected direct moving-bid grafts and retained small ammunition only at tick 1 below 1c, a large tick-1 bid only for 1c<p<2c, and a tick-1 ask inventory/time gate. The tuning paired gain is 2.0522924159996663 with standard error 0.5968385231758765, but the item explicitly ran no same-seed base holdout control, so its holdout is absolute performance rather than a paired generalization claim. Next, fit the stock target to observed demand while retaining cheap tick-1 acquisition unless new paired evidence supports moving it. [S11]

**combine-pair.** Tested latent belief alone and with overflow, then returned overflow with four FIFO slices and the latent implementation disabled by LAT = 0. Its reported +0.54 tuning gain becomes -0.004 on holdout, while belief variants did not beat overflow alone and increased arbitrage exposure. Next, estimate competitor queue depletion instead of a fixed $24 allowance, test ask-only overflow, and restore hard probe bounds before retrying latent belief. [S12]

**combine-three.** Retained an equal grid/latent belief blend with grid uncertainty, revised cash planning, and a latched scarce-stock two-cent join, rejecting the full belief swap and unconditional join. It reports the absolute holdout shown in the table but explicitly scored no holdout base control; one external source is flagged, so the combined mechanism's provenance is reported rather than independently confirmed and its small increments remain uncertain. Next, test queue waiting costs with removal controls on fresh seeds and examine the future value of cash. [S13]

**combine-all.** Tested six additions in sequence; only complete overflow passed the initial paired one-SE rule, and subsequent retained refinements restricted overflow to spreads 2-4, kept four FIFO slices, and expanded marginal quote capacity to 6000 while limiting main bids to 3000. Frozen tuning gain 1.5789076821303323 with paired standard error 0.384205139156741 did not yield a higher absolute holdout than the unchanged-base measurement, so this is not evidence that more grafts improve generalization. Next, estimate demand and queue depth online, price inventory by its probability of selling, and avoid selecting again on the disclosed holdout. [S14]

**combine-vote.** Ran synchronized base and late-game advisers with separate beliefs, electing late-game proposals in selected price/time regions only when their effective bid commitment did not exceed the base's. Tuning paired gain 0.969535002586166 has standard error 0.3909064991946845, but holdout gain 0.056340704598667295 has standard error 0.23682551783950723; one external source is flagged, so adviser provenance is an item claim and the local holdout supports near-neutrality. Next, isolate extra-stock and mid-price ask-collateral effects and test narrower guards and demand estimates on fresh seeds. [S15]

**demand-fit.** Fit retail arrival intensity and order notional jointly from capacity-conditioned own-order observations, using surviving orders to certify observations, then applied posterior demand to stock targets and marginal sizing. Its selected tuning paired gain is 1.9052810865828333 with standard error 0.8431690266178734 after extensive search, and the estimator retains finite-grid, probe and rounding approximations; one external source is flagged, so neither unbiasedness nor an independently verified public-leaderboard gain is established. Next, graft the conditional estimator into the strongest independently tested strategy and evaluate acquisition-cost-aware demand quantiles on new holdout seeds. [S16]

**tick1-jump-cost.** Restricted the resting tick-1 bid to within 3 one-step standard deviations on spreads 2-4 and below 1.7c on spread 1, reducing targeted jump-step losses. Reported paired tuning gain +0.59 +/-0.34 reverses to holdout -0.54 +/-0.43, so the item calls the overall result negative-to-null: saved jump cost also loses useful stock and retail sales. Next, avoid further blind window tightening and examine covered asks at ticks 3 to 9 using fill attribution and the acquisition cost of their stock. [S17]

**tick2-queue.** Expanded a capped tick-2 ask to covered stock before the spread-1 competitor joins, then preserved front quantity while the original model still wanted a positive ask. Holdout paired gain is 0.29151719602533344 with standard error 0.2504543285981268; the selected tuning gain is concentrated in a few paths, and unconditional expansion, splitting and expiry controls were less successful. Next, replicate on an untouched window before promotion, then model the useful lifetime of a front queue position. [S18]

**mid-jumps.** Added side/price-dependent weights only to modeled jump loss, with ask knots 1.0 / 2.5 / 2.0 and bid knots 0.7 / 1.0 / 1.0 at 5c / 50c / 95c. Reported paired tuning gain +0.760±0.251 shrinks to holdout +0.150±0.117 as retail gives back most arbitrage savings; diagnostics find middle-price asks more jump-exposed than bids, while crude flow estimates undershot demand. Next, diagnose flow observation bias before more risk-weight sweeps and test low-price stock-sale caps on more seeds. [S19]

**near-tick.** Changed only spread-1 runs: stop free-tick stock purchases and suppress a free-tick ask within 0.3 tick of estimated probability, retaining tick-1 acquisition. It reports spread-1 means 53.47 to 58.32 in tuning and 38.18 to 40.95 on holdout, but one external source is flagged and its small diagnostic samples do not prove the universal claim that no signal can improve mid-range probability knowledge. Next, test the two rules on spreads 2-4 and size tick-1 stock to measured retail flow, without assuming the untested spreads share the same economics. [S20]

## How to use the combined evidence

Prioritize cash planning for fresh paired replication, then investigate demand fitting, late-game stock controls and near-tick rules as separate candidates or carefully ablated grafts. Their mechanisms overlap in stock acquisition and cash allocation, so the results do not justify adding their gains; the ablation map directly demonstrates stock and spread-1 interactions. The weaker FIFO, constants and middle-jump candidates offer smaller reported effects, while belief replacement, unrestricted pair grafts, adviser voting and the six-way sequence provide no persuasive holdout lift here. This is a synthesis of reported evidence, not a newly tested combined strategy. [S4], [S6], [S7], [S8], [S9], [S10], [S11], [S12], [S13], [S14], [S15], [S16], [S17], [S18], [S19], [S20]

The oracle item discourages retail-parameter estimation under the base policy, whereas demand-fit and several next-step proposals favor it; this is not a resolved contradiction because knowing parameters in an otherwise fixed policy differs from changing stock targets and marginal demand modeling. Likewise, reducing jump exposure or deliberate overpayment can reduce retail profit: leaders-gap and tick1-jump-cost are explicit counterexamples to promoting a strategy solely for smaller arbitrage loss. [S2], [S5], [S16], [S17]

The measurement items agree that settlement mix, competitor spread and a concentrated profitable tail make independent 200-run scores noisy. Use per-seed paired differences, report uncertainty and outlier sensitivity, preserve the intended simulation distribution, and reserve a fresh evaluation window after any selection: the shared 30000–30599 block is now disclosed across this group and is no longer an untouched resource for the next wave. The unchanged-base measurements are repeated controls, not independent replications on different holdouts. [S1], [S3], [S4], [S9]

Do not normalize away reported discrepancies: several items reproduce base 102.331987422326 rather than 102.33197908899267, and combine-two-cent reports its own holdout base 96.00139032710622. The latter therefore has a small positive own-control delta despite its absolute candidate score lying below the common base control; late-game and combine-three explicitly lack their own holdout base controls. Sandbox outcomes also differ by item/runtime: failures before strategy loading in several reports coexist with successful checks elsewhere, so neither universal failure nor universal compatibility follows. [S6], [S8], [S11], [S13], [S14], [S15], [S16], [S19]

## Source reliability and provenance

The task explicitly flags one source each for leaders-gap, combine-belief, combine-two-cent, combine-three, combine-vote, demand-fit and near-tick as failing a later source check. It does not identify which URL or passage failed; those failures cannot be localized further from the supplied notice, and the affected claims are qualified above. The local experiment reports remain attributable to their item files but are not independent verification of their external citations; no missing evidence, measurements or source contents have been fabricated.

Each reference below names the supplied item result and its SHA-256; table metrics are its top-level `base_mean_edge`, `score`, `holdout` and `python_version`, and narrative findings come from `approach`, `tried` and `next`. Inputs were read from `.imd/reads/artifacts/item-KEY`; the canonical result paths below identify the contributor artifacts even when those files are not included in this merge bundle. Hashes identify exactly the source bytes used.

[S1]: ../../items/true-mean/result.json "item-true-mean"
- **S1 — item-true-mean**: `artifacts/goal/items/true-mean/result.json`; supplied read copy `.imd/reads/artifacts/item-true-mean`; SHA-256 `bce23e11838acc226f22f65932511adfa3165b372bf59d99afdfcb54912da59c`.
[S2]: ../../items/oracle-bound/result.json "item-oracle-bound"
- **S2 — item-oracle-bound**: `artifacts/goal/items/oracle-bound/result.json`; supplied read copy `.imd/reads/artifacts/item-oracle-bound`; SHA-256 `19276f30968b3eff40662ddb8b623e143bd7021922b72d33f93a3963c4ac6915`.
[S3]: ../../items/draw-variance/result.json "item-draw-variance"
- **S3 — item-draw-variance**: `artifacts/goal/items/draw-variance/result.json`; supplied read copy `.imd/reads/artifacts/item-draw-variance`; SHA-256 `9d2e52ba4d1ece29a5282a13b4b2ff16a1d230fa8e6bcc94379c6ff987ef991d`.
[S4]: ../../items/ablation-map/result.json "item-ablation-map"
- **S4 — item-ablation-map**: `artifacts/goal/items/ablation-map/result.json`; supplied read copy `.imd/reads/artifacts/item-ablation-map`; SHA-256 `9f2abf353a95d315d2ca1365aceccc8ba3ae8c21e3194881b15f980b5abbdca6`.
[S5]: ../../items/leaders-gap/result.json "item-leaders-gap"
- **S5 — item-leaders-gap**: `artifacts/goal/items/leaders-gap/result.json`; supplied read copy `.imd/reads/artifacts/item-leaders-gap`; SHA-256 `64c3834962a807c9928f787e7d10462124376ba048c75536ebc78e25f9436229`.
[S6]: ../../items/combine-belief/result.json "item-combine-belief"
- **S6 — item-combine-belief**: `artifacts/goal/items/combine-belief/result.json`; supplied read copy `.imd/reads/artifacts/item-combine-belief`; SHA-256 `e0ef348d5989364eaa0ec9ba378232df6a50b3de691ec4377d61bdb0438e9120`.
[S7]: ../../items/combine-overflow/result.json "item-combine-overflow"
- **S7 — item-combine-overflow**: `artifacts/goal/items/combine-overflow/result.json`; supplied read copy `.imd/reads/artifacts/item-combine-overflow`; SHA-256 `e030317a9b88b7935b96704d7d4577d8d459b9e2004f8d82318c76795c6a51f7`.
[S8]: ../../items/combine-two-cent/result.json "item-combine-two-cent"
- **S8 — item-combine-two-cent**: `artifacts/goal/items/combine-two-cent/result.json`; supplied read copy `.imd/reads/artifacts/item-combine-two-cent`; SHA-256 `741d1df09e300655960b5c8c4b4fdd26aba73faf539ea90f6b5f2e8d64ed924e`.
[S9]: ../../items/combine-cash/result.json "item-combine-cash"
- **S9 — item-combine-cash**: `artifacts/goal/items/combine-cash/result.json`; supplied read copy `.imd/reads/artifacts/item-combine-cash`; SHA-256 `ecbc9087b4f238fc1b5b0d52f55af073f676c038bfb6756172b1e51480ac1741`.
[S10]: ../../items/combine-constants/result.json "item-combine-constants"
- **S10 — item-combine-constants**: `artifacts/goal/items/combine-constants/result.json`; supplied read copy `.imd/reads/artifacts/item-combine-constants`; SHA-256 `0a7cab773a6c37f78cc30f55f956b0ee39e25495f6dbcbdc8effb55e12f3f119`.
[S11]: ../../items/combine-late-game/result.json "item-combine-late-game"
- **S11 — item-combine-late-game**: `artifacts/goal/items/combine-late-game/result.json`; supplied read copy `.imd/reads/artifacts/item-combine-late-game`; SHA-256 `1905c7e39765aa4e1d0dbb2f5cb731a6cf724fc614f63987386857acc332dcb3`.
[S12]: ../../items/combine-pair/result.json "item-combine-pair"
- **S12 — item-combine-pair**: `artifacts/goal/items/combine-pair/result.json`; supplied read copy `.imd/reads/artifacts/item-combine-pair`; SHA-256 `e645b386d55814ff94d542561e82f6aba887947070c1189f25f24fa15881a147`.
[S13]: ../../items/combine-three/result.json "item-combine-three"
- **S13 — item-combine-three**: `artifacts/goal/items/combine-three/result.json`; supplied read copy `.imd/reads/artifacts/item-combine-three`; SHA-256 `4fcbb41c04be5f2db1aeb6fd36176bfbb7b1cd959d24434f87d73cc51e1eb100`.
[S14]: ../../items/combine-all/result.json "item-combine-all"
- **S14 — item-combine-all**: `artifacts/goal/items/combine-all/result.json`; supplied read copy `.imd/reads/artifacts/item-combine-all`; SHA-256 `3d1a891acf1dea4a5a217530623229881ac45f2d8c237e7063d0c672803795be`.
[S15]: ../../items/combine-vote/result.json "item-combine-vote"
- **S15 — item-combine-vote**: `artifacts/goal/items/combine-vote/result.json`; supplied read copy `.imd/reads/artifacts/item-combine-vote`; SHA-256 `134acaa04b74af5218d392bf5af04f3ef74e95117f26e2e1b681a037d3fd7125`.
[S16]: ../../items/demand-fit/result.json "item-demand-fit"
- **S16 — item-demand-fit**: `artifacts/goal/items/demand-fit/result.json`; supplied read copy `.imd/reads/artifacts/item-demand-fit`; SHA-256 `12a3e68330f2c22c6a1d148519d91ea4dd37431d724fe1f9b3d19882d463c4d8`.
[S17]: ../../items/tick1-jump-cost/result.json "item-tick1-jump-cost"
- **S17 — item-tick1-jump-cost**: `artifacts/goal/items/tick1-jump-cost/result.json`; supplied read copy `.imd/reads/artifacts/item-tick1-jump-cost`; SHA-256 `1176ed1afc412337527d637b8c39079062c7c285edfbdd2c3d81294738981c02`.
[S18]: ../../items/tick2-queue/result.json "item-tick2-queue"
- **S18 — item-tick2-queue**: `artifacts/goal/items/tick2-queue/result.json`; supplied read copy `.imd/reads/artifacts/item-tick2-queue`; SHA-256 `1c6d2be40f07a654f8b788d08acb427bae801997ad17e654849155c922d857e6`.
[S19]: ../../items/mid-jumps/result.json "item-mid-jumps"
- **S19 — item-mid-jumps**: `artifacts/goal/items/mid-jumps/result.json`; supplied read copy `.imd/reads/artifacts/item-mid-jumps`; SHA-256 `4215b9d8a4f6ca65dd31fa59d851884636090cea282b6dd151218cd15fbe3ea5`.
[S20]: ../../items/near-tick/result.json "item-near-tick"
- **S20 — item-near-tick**: `artifacts/goal/items/near-tick/result.json`; supplied read copy `.imd/reads/artifacts/item-near-tick`; SHA-256 `4dffd5e77678c33d58a008d05bbe01d0be4b201d79fd07097949a9d1b7610382`.