Quantitative empirical research on judicial behaviour is not yet as developed in the United Kingdom as it is in some other jurisdictions. Perhaps, as Chris Hanretty suggests, this is partly because ‘investigating the non-legal factors that determine judges’ decisions might … be taken as a rather bad-mannered thing to do’ (265) (although, as Hanretty himself shows, quantitative research may focus on legal factors as well). Another factor is the preference among legal academics for doctrinal or philosophical over empirical research methods; which doesn't explain the relative lack of interest among academics in other disciplines. Whatever the reasons, the result is that many interesting research questions that could be addressed through quantitative studies remain to be tackled. In his valuable and pioneering book, A Court of Specialists: Judicial Behavior on the UK Supreme Court, political scientist Hanretty sets out to answer some of those questions. Hanretty's monograph is unabashedly quantitative in orientation, in contrast to previous empirical studies of the UK's highest court based on qualitative data such as interviews or analysis of judges’ notes. Without attempting in this review to offer a direct methodological defence, I will disclose my bias in favour of a quantitative approach. A Court of Specialists directly addresses potential resistance from legal readers and does so effectively, especially by careful attention to the limitations of the conclusions that it draws – which are mostly ‘myth-burnishing’ rather than ‘myth-busting’ (265). For readers familiar with comparable studies of high courts, it is striking how attractive the Supreme Court ends up looking in A Court of Specialists. Hanretty found that ‘the law strongly constrains … outcomes’ in the Supreme Court (271). Unlike studies of judiciaries in other jurisdictions (especially on the United States Supreme Court), Hanretty's best predictive model of a case outcome in the UK Supreme Court relies chiefly on a legal factor, the ‘opinion below’. Other myth-burnishing conclusions include that the Court appears to work hard to avoid forming panels dominated by judges who share the view of the presiding judge and that cases selected to be heard by the Court (granted permission to appeal) genuinely raise arguable points of law of general importance. Hanretty was able to reach those conclusions because he took the law seriously, striving to identify the best available quantifiable proxies for legal factors in adjudication. He considered: the balance of opinion of judges below, the area of law, specialisations of Supreme Court judges and legal importance of an appeal. The first and the last of those are particularly interesting. Hanretty operationalised ‘opinion below’ as a ‘simple ratio of the number of judges who have heard in favor of the appellant relative to all judges who have heard the case’ (35). He found that the magnitude of the ratio of judge-votes below in the appellants’ favor is by far the strongest predictive factor of a final outcome in the Supreme Court (260). The ratio so constructed does appear to be an eminently reasonable proxy for what the courts below think the law requires in each case. Of course, it is not perfect. For any case, an observer might think that several judges who heard the case below were mostly legally mistaken. But to say that this is true with enough regularity to make this measure a bad proxy of (at least reasonable) judgment of what the law requires would be tantamount to rather extreme scepticism about judicial competence. Even more so given how often the Supreme Court agrees with opinion below. What Hanretty does not attempt, but would be interesting to see in future research, is a more detailed breakdown of opinion below to measure the predictive force of judge-votes from specific tribunals: the High Court, the Court of Appeal, the Upper Tribunal, etc. A second notable parameter, legal importance of an appeal, Hanretty ‘measured by the number of generalist law reports that reported the case in the originating court’ (46). Given the number of cases under analysis (all 674 cases decided by the UK Supreme Court between its creation in 2009 and September 2017), I empathise with the decision not to go for a more complicated proxy, for example including press coverage or social media mentions. In line with what might be expected, legal importance operationalised this way correlated positively with grants of permission to appeal by the Supreme Court and with increased panel sizes. The ‘specialists’ of the book's title references the remaining two of Hanretty's four highlighted predictive variables: the area of law engaged by an appeal and judges’ specialisations. Hanretty found that specialisation is the strongest factor explaining which judges are chosen to hear an appeal and that this effect is ‘larger in more important cases’ (130). The same effect holds for which judge is selected to write the lead opinion, but with an inverse relation: in more important cases the effect is smaller (164-165). Conclusions regarding who writes substantive opinions other than the lead opinion are not as statistically robust, but they point in the same direction of specialist influence (165). Given that the Supreme Court hears most cases in panels of five and that lead judgments predominate, the influence of specialisation effects on the Court's jurisprudence justifies Hanretty's framing and title choice. In addition to strictly legal influences, Hanretty also examined ‘organisational’ and ‘political’ factors. The first variable measured legal importance (as a factor for outcomes other than grants of permission to appeal) and judges’ workload; the second tracked the extent of agreement between judges and the type of party to the litigation. As anticipated, importance correlated positively with panel size and central government tends to have it easier with permissions to appeal and its participation increases the likelihood of a case being heard by a larger panel. However, the effects of workload and agreement between judges are more surprising. As Hanretty phrased it: ‘[t]here is, in the way the Supreme Court allocates judges to panels, a Matthew effect, whereby those with high workloads find that they are given more still’ (130). And somewhat counter-intuitively, ‘those who agree with the President of the court are less likely to be empaneled than those who disagree with the President’ (130), though this effect is not large. Chapter 8, on ‘Political Patterns of Dissent’, places Supreme Court judges on a left-right political spectrum according to what outcome they favoured in every case. These designations are inevitably controversial. For example, a ‘win for the individual’ in a criminal law case was coded as a ‘left-wing outcome’, whereas in a public law case between a public authority and a private company a ‘win for public authority’ was coded this way (208). Notably, a ‘win for the party alleging human rights infringement in human rights cases’ was always coded as a left-wing outcome (208). As a result, Lord Kerr took the crown of the most left-wing judge with Lady Hale and Lord Toulson also on the podium (214). Lord Brown, Lord Walker and Lord Rodger were closely tied at the other pole of the political spectrum, with Lord Sumption and Lord Reed closer to the centre. This analysis is not in fact as controversial as it might appear. First, the results are well-aligned with the common opinion of experienced Court-watchers and the results of previous studies (eg Arvind and Stirton, ‘Legal Ideology, Legal Doctrine and the UK's Top Judges’ [2016] Public Law 418). Secondly, and as Hanretty stresses, this analysis was based only on dissents, which are relatively uncommon in the Supreme Court. These findings cannot therefore be interpreted as ‘generally important’ (237). From a public lawyer's perspective, A Court of Specialists could have addressed more robustly the influence of human rights on the Court, a topic which seems to dominate public and academic debates about the institution. Alan Paterson suggested that, during the last years of the House of Lords, the ‘“A” team for Human Rights cases’ was responsible for an increased percentage of human rights claims being granted leave to appeal (Final Judgment: the Last Law Lords and the Supreme Court, Hart Publishing, 2013, 68). Responding to Paterson's suggestion, Hanretty found ‘no link between panel specialization and PTA [permission to appeal] success’ (275). This may be true, but it does not address Paterson's specific point about human rights. Hanretty did not analyse the judges’ human rights ‘orientation’ (for want of a better word) or its effects on grants of permission. He included the human rights aspect only when analysing political patterns of dissent. This is an unfortunate lost opportunity. It would be very interesting to know whether, for example, a permission panel that included Lady Hale or Lord Kerr was more likely to grant leave to appeal on a human rights issue. It would also be interesting to investigate potential correlations between human rights-favouring final outcomes and particular serving Justices, also in unanimous cases. In his suggestions for the direction of future quantitative research on UK courts, Hanretty underscores the value of studying appellate courts below Supreme Court level. I don't dissent, but I think there is still plenty to learn from further quantitative investigations of UK Supreme Court practice.