AstroGuide:Portfolio Chris Hollander’s Astrophotography

AstroGuide destinations

Recommendation engine transparency

Where does the advisor actually send you?

This report runs every telescope pairing through every possible answer pattern, then measures where the recommendation lands. It is a behavioral audit of the engine—not a market survey, product ranking, or claim that every scenario is equally likely in real life.

13,440complete scenarios
15selectable telescopes
4platform ecosystems
64answer patterns per pair

The fairest platform comparison

Cross-platform acceptance rate

For each current platform and candidate platform, this measures how often the final recommendation comes from the candidate’s ecosystem. It removes the distortion caused by platforms having different numbers of models in the advisor.

Row platform currently owned Column platform being considered Cell candidate platform recommendation rate
Current ↓ / Candidate → Seestar DWARFLAB Vaonis Unistellar
Seestar 4 models — 68.1% 697 of 1,024 49.2% 378 of 768 53.7% 550 of 1,024
DWARFLAB 4 models 77.2% 791 of 1,024 — 45.4% 349 of 768 50.4% 516 of 1,024
Vaonis 3 models 66.1% 508 of 768 56.9% 437 of 768 — 42.7% 328 of 768
Unistellar 4 models 69.7% 714 of 1,024 64.8% 664 of 1,024 35.7% 274 of 768 —

A 73% cell does not mean that platform is “73% better.” It means that, across the equally weighted telescope pairs and answer patterns represented by that cell, the engine selected a telescope from the candidate platform 73% of the time.

Confusion-matrix view

Where recommendations land

Rows are the platform currently owned; columns contain the final recommended telescope’s platform. Percentages are normalized within each row because platforms contribute different numbers of current models.

Current ↓ / Recommended → Seestar DWARFLAB Vaonis Unistellar
Seestar 3,584 runs 54.7% 1,959 scenarios 19.4% 697 scenarios 10.5% 378 scenarios 15.3% 550 scenarios
DWARFLAB 3,584 runs 22.1% 791 scenarios 53.8% 1,928 scenarios 9.7% 349 scenarios 14.4% 516 scenarios
Vaonis 2,688 runs 18.9% 508 scenarios 16.3% 437 scenarios 52.6% 1,415 scenarios 12.2% 328 scenarios
Unistellar 3,584 runs 19.9% 714 scenarios 18.5% 664 scenarios 7.6% 274 scenarios 53.9% 1,932 scenarios

The diagonal includes both “keep your current telescope” and “choose another model on the same platform.” It should not be interpreted as a pure platform-loyalty score.

Inside the diagonal

Keep, switch models, or switch platforms?

This separates an exact “keep what you own” result from a move to another model in the same ecosystem and a move across platforms.

ZWO

Seestar

39.8%Keep the current telescope

1,427 scenarios

14.8%Choose another model on the same platform

532 scenarios

45.3%Move to another platform

1,625 scenarios
DWARFLAB

DWARFLAB

40.3%Keep the current telescope

1,443 scenarios

13.5%Choose another model on the same platform

485 scenarios

46.2%Move to another platform

1,656 scenarios
Vaonis

Vaonis

49.5%Keep the current telescope

1,331 scenarios

3.1%Choose another model on the same platform

84 scenarios

47.4%Move to another platform

1,273 scenarios
Unistellar

Unistellar

50.6%Keep the current telescope

1,814 scenarios

3.3%Choose another model on the same platform

118 scenarios

46.1%Move to another platform

1,652 scenarios

Methodology and limitations

Exactly what this does—and does not—measure

Platform capability ratings explain the ownership experience. They do not contribute points to Upgrade Advisor recommendations.

  1. 01

    Enumerate every ordered telescope pair

    Each of the 15 telescopes is treated as the current telescope and compared with every other telescope: 15 × 14 ordered pairs.

  2. 02

    Run every answer pattern

    The advisor currently asks 6 two-choice questions, producing 64 possible answer patterns for every pair. Together, that creates 13,440 complete runs.

  3. 03

    Record the final recommended telescope

    The report groups that telescope by platform, preserving both exact-retention and same-platform model changes. Cross-platform alternatives are restricted to the platform the user selected as the candidate.

  4. 04

    Weight everything equally

    No usage analytics, sales data, or inferred customer profile is used. Rare and common answer patterns receive identical weight, as do every current and candidate telescope pairing.

This is not user telemetry

The report describes possible engine behavior. It does not claim that real visitors choose platforms or answers in these proportions.

The catalog affects the matrix

Platforms with more selectable models create more pairings. The acceptance-rate matrix compensates for that; the destination matrix deliberately shows the complete engine surface.

Non-current models remain included

Discontinued and preorder telescopes remain selectable because the advisor supports used-market comparisons and evaluation of announced hardware.

Capabilities are explanatory

Mosaics, citizen science, digital eyepieces, editing tools, and other platform features appear in comparison tables but never contribute recommendation points.