277 A.3d 991
Md.2022Background
- Two victims were killed with a 12‑gauge shotgun in Anne Arundel County; nearby security cameras captured a person carrying what appeared to be a shotgun, but the face was indiscernible.
- A reverse‑projection photogrammetry analysis by the FBI concluded the person’s height (to top of headwear) was approximately 5'8" ± 0.67", but the FBI report warned several variables could make the true uncertainty “significantly greater.”
- Matthews was measured by police at about 5'9" and was later indicted; defense moved to exclude the FBI expert under Md. Rule 5‑702 and 5‑403 (and Frye‑Reed then in effect).
- At the pretrial hearing and at trial the FBI analyst explained her reverse‑projection method, calculated a quantifiable error for some factors, but conceded she could not scientifically quantify other variables (terrain, footwear, headwear fit, body position, image resolution). Defense cross‑examined and presented a contrary photogrammetry expert.
- Trial court admitted the expert (finding methodology reliable and testimony helpful), the jury convicted Matthews, the Court of Special Appeals reversed (finding an analytical gap), and the Court of Appeals reversed that reversal and reinstated convictions.
Issues
| Issue | State's Argument | Matthews' Argument | Held |
|---|---|---|---|
| Admissibility under Md. Rule 5‑702 (reliable methodology; analytical gap) | Photogrammetry (reverse projection) is a reliable technique; the expert applied it to adequate data; unknown additional uncertainty goes to weight, not admissibility. | Unknown/unquantified variables produced an unknown margin of error that created an analytical gap making the opinion unreliable and inadmissible. | Court of Appeals: No abuse of discretion in admitting the testimony—methodology reliable, no analytical gap, unquantified uncertainty goes to weight and cross‑examination. |
| Whether inability to quantify all sources of uncertainty is per se grounds to exclude | Not per se; Daubert/Rochkind factors are flexible and judge has broad discretion—lack of a quantified total error can be explored at trial. | Where core variables are unknown and unquantifiable, the expert cannot reliably bridge data to conclusion; exclusion required. | Court: Not per se; trial court should assess whether uncertainty reflects methodology flaws or only affects conclusions; here it affected weight, not admissibility. |
| Md. Rule 5‑403 balancing (probative value vs. unfair prejudice/misleading jury) | Expert testimony was probative on identity and corroborated other testimony; limitations were disclosed and available for cross‑examination. | The apparently precise estimate (5'8" ± .67") was misleading given the unquantified true error, so probative value was minimal and risk of misleading/prejudice high. | Court: Trial court acted within discretion—probative value not substantially outweighed by unfair prejudice; limitations were explicit and contestable at trial. |
Key Cases Cited
- Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579 (U.S. 1993) (establishes reliability inquiry for admitting expert scientific testimony)
- Gen. Elec. Co. v. Joiner, 522 U.S. 136 (U.S. 1997) (trial court may exclude testimony where there is an analytical gap between data and opinion)
- Kumho Tire Co. v. Carmichael, 526 U.S. 137 (U.S. 1999) (Daubert gatekeeping applies to technical and other specialized expert testimony; broad trial‑court discretion)
- Rochkind v. Stevenson, 471 Md. 1 (Md. 2020) (adopts Daubert framework under Md. Rule 5‑702; reliability is flexible, no single dispositive factor)
- Savage v. State, 455 Md. 138 (Md. 2017) (discussion of analytical‑gap concept under Md. Rule 5‑702)
- Blackwell v. Wyeth, 408 Md. 575 (Md. 2009) (methodology must be coupled with proper analysis to avoid an analytical gap)
- Matthews v. State, 249 Md. App. 509 (Md. Ct. Spec. App. 2021) (intermediate appellate decision excluding expert testimony as suffering an analytical gap)